PP3: SOCIO-ECONOMIC MODEL AND COMMUNITY IMPACT SIMULATORS

Methodology:

1. Conceptual model (Months 1 to 6):

There is, at present, no model that connects the sectors of interest (housing, crime, health, employment, SMEs, and accident/rescue incidents), let alone one that connects these sectors to extreme weather events. So an important precursor to the development of the EWESEM (Extreme Weather Events Socio-Economic Model) is the construction of a robust conceptual framework to underpin the empirical analysis. This conceptual framework must be amenable to mathematical representation but also based on sound theoretical understanding of the links between variables. Our theoretical model will be informed by the neighbourhood effects literature in which advances have recently been made in developing dynamic models at the local level (Galster et al 2007; Galster 2007) and also by the existing literature associated with the individual sectors. For example, Hwan and Quigley (2006) construct a system of equations connecting housing vacancy rates with house price and new construction. Bramley and Leishman (2005) construct a sub-regional socio-economic five equation panel model of house prices, in-migration, out-migration, new build and vacancies. Lim and Galster (2007) extend and synthesise the deterrence (Becker 1968, Stigler 1970) and social-interaction models (Murphy et al 1993, Glaeser et al 1996) of crime to a microeconomic model of year-to-year changes in crime rates that incorporates endogenous relationships between the recruitment of criminals and deterrent effects spawned by responses of neighbourhood residents and/or police.

2. Data collation and cleaning, Data summaries and description (Months 1 to 12):

The implication of constructing a community level model (as opposed to a regional or macro one) is that the datasets for each of the core modules (housing, crime, health, employment, SMEs, and accident/rescue incidents) are potentially very large. Collating and cleaning these data will therefore entail a significant amount of manpower and expertise, and will require a researcher with appropriate experience in manipulating large social science datasets. Once cleaned and collated, we shall produce summary tables at levels of aggregation necessary to meet the confidentiality requirements of the data suppliers, but also at a level that will be of use to Programme Packages 1 and 2.

3. Construction of derived variables (Months 7 to 12):

To be methodologically robust, many of the required variables will require a complex derivation process. For example, simply developing an appropriate measure of local house prices will involve a significant degree of modelling in order to correct for selection bias, mix adjustment and missing data. Without such corrections, observed house price changes may be spurious. An apparent increase in average house prices may, for example, simply reflect changes to the mix of properties coming onto the market, rather than genuine changes to the value of the housing stock in the locality. (see Gatzlaff and Haurin 1994; Fik et al 2003; Pryce and Evans 2007; Pryce and Mason 2006).

4. Models of Individual Sectors and an Integrated Model (Months 7 to 30):

We shall first construct separate models of the impact of extreme weather events on each of the individual sectors (this is the most appropriate way of tackling the complexity of each of the sectoral impacts). These models will then be trimmed and integrated into a single multiple-equation model that will form the basis of the Community Impact Simulators and the scenario building.

5. Community Impact Simulators (Months 1 to 30):

Once the parameters capturing the relationship between the variables in our model have been estimated, they can be used to develop a simple simulation tool that allows users to run a variety of Type I scenarios (Type I scenarios are those that consider the impacts of extreme weather events happening in the immediate future). It is anticipated that these Simulators will have a GIS interface and will therefore be developed in conjunction with the GIS Programme Package (PP5: Mapping Extreme Weather Events and Impacts), and informed by stakeholder engagement (facilitated by PP1, PP2, PP3, PP6). We would also seek to explore the feasibility of integrating the Community Impact Simulators with existing models that have been developed in particular sectors (such as the CLG FSEC model for the Fire and Rescue service, ODPM 2006, and the CLG Affordability Model for the housing sector, Meen 2005).

6. Type II _What if? Scenarios (Months 24 to 30):_

Drawing on existing research (such as the BESEECH and CLG Affordability projects) we will establish a range of trajectories for the socio-economic variables in the model. This will allow us to develop a limited number of Type II scenarios exploring the impact at particular points in the future of extreme weather events as specified by the Extreme Weather Events Estimator (PP4). These scenarios would also be represented in graphical form via collaboration with PP5.

7. Writing-up (Months 31 to 36):

We shall endeavour throughout the project to maintain good written records of all methods, outputs, models, simulators and results. These will then be drawn together in a dedicated writing-up phase in the final three months of the project to produce a technical report on the Socio-Economic Model and Community Impact Simulators and also a policy (non-technical report). These will form the basis of future published outputs, such as journal articles, books, and briefing notes developed both within PP3 and collaboratively with other Programme Packages.

Deliverables:

• Modelling tools and prototype interactive software-based Community Impact Simulators to facilitate decision making and the application of these resources to particular "What if?" scenarios (including a range of assumptions for weather patterns, coping/adaptation strategies and other socio-economic variables). (Draft versions supplied to particular stakeholders in Months 6, 21 and 30; final version in Month 36)
• Maps and other graphical representations of the community impacts of extreme weather scenarios (Draft in Month 30 and Final version in Month 36).
• A Technical Report on the EWE Socio-Economic Model and the Community Impact Simulators, as an individual report and/or as part of a larger collaborative report produced in conjunction with the other PPs(Draft in Month 36) .
• A short Policy Report (non-technical) on the social, economic and policy implications of the EWE Socio-Economic Model and the Community Impact Simulators, as an individual report and/or as part of a larger collaborative report produced in conjunction with the other PPs(Draft in Month 36).
• Innovative framework that integrates socio-economic modelling with climate models and weather event generators (to be summarised in above reports, Drafts in Month 36).
• Industry briefing papers, articles in industry periodicals, and non-technical reports.
• A dedicated website from which research outputs and interactive materials can be accessed, and through which feedback in the longer-term can be collated (to be developed iteratively throughout the project, completed by Month 36).
• Academic outputs in the form of journal articles and edited books, both of PP3 specific material, and of cross-PP results.

Innovation/Timeliness:

To our knowledge this will be the first attempt to develop a community level model that integrates the health, employment, housing, crime, emergency services sectors with long-term predictions of climate change and extreme weather event scenarios, and that does so in a way that captures both the immediate and lagged effects. At present, long-term projections of the housing market (the CLG Affordability Model, Meen 2005), for example, and decisions on the optimal location of new housing take no account of climate change and the implications for flood risk. Both the tangible outputs (prototype decision making tools) and the conceptual and methodological frameworks that underpin those outputs and link socio-economic and weather models, represent significant innovations.

Management of Project

A 3 year Research Programme is proposed. Overall Programme management will take place as a separate activity in PP6/6. The Programme Package 3 will be managed by the PI Pryce to ensure that key outputs are delivered at the appropriate milestones to the required quality; EPSRC is informed of progress and is consulted should opportunities for innovation arise or unforeseen deviations emerge. The CWEWE Programme Management Group, made up of individual PP PIs, will also meet 6 monthly (see Gantt chart for details). PI Pryce will engage with the CWEWE Programme Manager to ensure that appropriate links and collaborative arrangements are in place with other PPs, including both scientific and managerial aspects .

Outputs to Other Programme Packages (PP1, PP2 and PP5) and to Stakeholder Groups where appropriate via PP2 and PP1 (see deliverables above and table below):

Note that codes in the second column refer to points on the PP3 GANTT Chart.

Inputs from other packages: to come via feedback on these outputs (particularly the conceptual model and Community Impact Simulators), and inputs on EWEs provided by PP4.

References:

Bramley, G. and C. Leishman (2005). "Planning and Housing Supply in Two-Speed Britain: Modelling Local Market Outcomes". Urban Studies, Vol. 42, No. 12, 2213-2244 .

Fik, T.J., Ling, D.C., Mulligan, G.F. (2003) Modelling spatial variation in housing prices, Real Estate Econ., 31 (4): 623-646

Galster, G. (2007) Neighbourhood and Social Mix as a Goal of Housing Policy: A Theoretical Analysis, European Journal of Housing Policy, 7(1), 19-44.

Galster, George, Jackie Cutsinger, and Up Lim. 2007. Are Neighborhoods Self-Stabilizing? Exploring Endogenous Dynamics. Urban Studies 44: 167-185.

Gatzlaff, D.H. and D. R. Haurin (1994) Measuring changes in local house prices, Journal of Urban Economics, 35, 221-244.

Glaeser, Edward L., Bruce Sacerdote, and José A. Scheinkman. 1996. Crime and social interactions. Quarterly Journal of Economics 111:507-548.

Lim, U. and Galster, G. (2007) The Dynamics of Neighborhood Property Crime Rates, Working Paper, Wayne State University, Department of Geography & Urban Planning.

Murphy, Kevin M., Andrei Shleifer, and Robert W. Vishny. 1993. Why is rent-seeking so costly to growth? American Economic Review 83:409-414.

Pryce, G. and Evans, G. (2007) Identifying Submarkets at the Sub-Regional Level in England, Department of Communities and Local Government: London.

Pryce, G. and Mason, P. (2006) Which House Price? Finding the Right Measure of House Price Inflation for Housing Policy - Technical Report, Office of the Deputy Prime Minister, ISBN: 05 ASD 03771/a.

Stigler, George J. 1970. The optimum enforcement of laws. Journal of Political Economy 78:526-536.

Meen, G. (2005) Affordability Targets: Implications for Housing Supply, Office of the Deputy Prime Minister, London.

ODPM (2006) Potential further developments of FSEC, Office of the Deputy Prime Minister, London.