Reviewed by
Sven Stöwhase
Fraunhofer-Institute for Applied Information Technology, FIT, Microeconomic Modelling Group, MIKMOD
This book is based on selected papers from the 1st General Conference of the International Microsimulation Association, held in 2007 in Vienna. The term "New Frontiers" in the title already gives us a first impression of its content. Rather than focussing on formal or theoretical advances in modelling, this book aims to illustrate current developments and advances in microsimulation modelling that are based either on improvements in data quality or policy-relevant research questions emerging recently. In particular, advances in spatial modelling, behavioural modelling, and in micro-macro linkages, as well as new developments of microsimulation models for demographic issues, such as fertility decisions and pension policies, are presented. Consequently, the selection of papers and their results are quite heterogeneous, thereby allowing for a general overview on recent developments. On the other hand, each single contribution on its own is very specialised and may be of interest only for a smaller group of "expert" readers.The structure of the book follows the above topics. The editor's introduction, which briefly reviews the contributions, is followed by a section on spatial modelling, with four contributions that deal with issues such as spatial migration, the spatial measurement of poverty, household shopping and spending behaviour. In addition to partial results, the corresponding data bases, and simulation approaches - with all their problems - are presented as well. The selection of contributions in this section, as well as in the other sections of the volume, shows thematic but also regional differences, since simulation models come from the UK, the Netherlands and Australia.
The focus of the next section is on modelling labour supply decisions. This has become a topic of growing interest in the last years, especially when it comes to the evaluation of tax reforms and changes in welfare programs. The perhaps most interesting contribution in this section is devoted to labour market issues and investigates to what extent the labour demand side can be modelled on the micro level, offering solutions for the matching problem with the supply side. Another interesting chapter deals with the theme of policy swapping in the context of the (well known) EUROMOD project. Here, a case-study from Southern European countries is presented that gives a brief overview about the potentials of such analysis for policy purposes.
The following section focuses on demographic issues. It deals both with comparatively simple models estimating the fiscal effects of pension reforms and more complex dynamic models on lifetime earnings and pensions. This section demonstrates the current limits of microsimulation. One example is the modelling of fertility timing and fertility decisions which would have needed more information on social interactions which strongly influence fertility. Despite these limits, the fact that this section with nine contributions is far more comprehensive than the others may reflect the editors' view of the (future) significance of demographic issues for microsimulation.
The final section is primarily dedicated to the linkage of micro- and macro-models. Examples provided here comprise models of tax-transfer policies, health policy, and immigration, where the results obtained on the micro level have a feedback on macroeconomic variables and vice-versa. In contrast with traditional microsimulation, these types of models increase the validity of the results, since they allow us to derive total effects instead of partial effects.
Overall, the contributions presented in this volume give a very good overview of the state-of-the-art in microsimulation modelling. The volume may be useful for at least two different tasks. Firstly, it could be a reference for those already working in the area of microsimulation who are searching for guidance towards specific topics. Secondly, it could be useful for those interested in building a microsimulation model, since the selection of papers provides a broad overview of the scope but also about the potential pitfalls associated with model building.
The only critic I have is that, while this volume is almost too broad in scope, some papers seem to be written with a very small readership in mind. Due to shortage of space, the modelling techniques and data descriptions are often described in a superficial way that does not make the reader understand the methods used in full detail. Furthermore, the focus is sometimes almost exclusively on model results, whereas aspects of the modelling approach are left too much on the background. As such, these contributions would have been more appropriate for specialised journals.
Despite these weak criticisms, I must say that this book is worth reading for the JASSS audience, and I must thank the editors for their excellent selection of contributions. This book will become definitively one of my first references in the future and I guess so it will be for anyone interested in microsimulation.
Return to Contents
of this issue
© Copyright Journal of Artificial Societies and Social Simulation, 2010