Reviewed by
Andreas Koch
University of Salzburg
For readers who love to have a single word straightforwardly summarizing the value of a book, I would say that this book by Iannis M. Ioannides is simply "comprehensive". This is a 450 pages book that develops the idea of social interactions as a crucial factor to explain economic relationships and follows a strong spatial understanding of these both in terms of geographical and social spaces and as topological spaces, e.g., using graph and network theories. Ioannides covered all relevant literature comprehensively and extensively, with recent references as well as seminal historical works by economists, economic geographers, and - even if less extensively - urban and social geographers. This makes the book a valuable reference for everyone involved in these fields of research. After introducing the theory and empirics of social interactions, the book zeroes in on the location decisions of individuals and firms, focusing on endogenous, exogenous, and contextual problems of neighborhoods and agglomeration effects. The remaining chapters touch upon the next higher scales of urban infrastructure and intercity trade. Each chapter, first, provides the reader with a brief overview of the issues at hand and then presents their mathematical formalization. As Ioannides is strongly sensitive towards contextualized readability, the chapters are not built as independent and mutually isolated fragments. By repeatedly referencing to models and authors used in previous chapters, mutual references between chapters are possible that allow the reader to identify common rules across all the applied models. This stimulates the reader to transfer the models presented in the book into his/her own research context.
An additional benefit is that chapter 2 ("Social Interactions: Theory and Empirics") includes two comprehensive appendices, one explaining basic facts of graph and network theory and the second describing data sources used throughout the book. At the end of chapters 6 and 9, an annotated overview of literature is presented as core references in the prior chapters (chapters 3 to 6 and 7 to 9, respectively).
From neighborhoods to nations is an examination of the social domains within economic activities across all levels of social and spatial scales. Therefore, another proper title could have been "from local to global" or "from micro to macro". Ioannides advocates a more advanced and comprehensive consideration of social theory and methodology in the applied empirical work of econometricians. His main aspirations is summarized in his same words as follows: "the book is designed to enrich our set of metaphors for understanding and modeling the fabric of communities, their neighbors, and their consequences for studying larger regional and national economies", "[m]y overarching theme [...] is proximity in all of its dimensions and its impact on interactions among individuals and firms in society and in the economy", and "[m]y goal is to emphasize that our knowledge of social interactions rests on data, on the empirical findings that derive from them, and on the applied economics that made those findings possible" (p. xi-xii).
Although someone could view the ambition of this book as too excessive, Ioannides took a pragmatic epistemological path by restricting his attention to quantitativey tractable concepts, such as "fabric", "proximity", and "data", e.g., social theory as synonymous with computational sociology, linked with economic (and partly social) geography. Remarkably, the book does not contain concrete examples of agent-based models, although typical social simulation topics, such as Schelling’s segregation model, are included. This is not necessarily a disadvantage, as the book provides readers with a plethora of mathematical models which help us to understand the formalization and operationalization of complex social phenomena.
The recurrent theme of the whole book is the differentiation of social effects: first, endogenous social effects which are determined by individual actions and expectations on actions in a social community. The action of each individual is always embedded into and associated with sets of actions of others which the former is socially related to. Secondly, exogenous social effects are relevant through personal characteristics of other individuals in a specific social community. Thirdly, correlated social effects refer to social institutions or socio-cultural environments that socially frame people’s behaviour. Despite the fact that all these three dimensions are plausible and can be easily discriminated theoretically, treating them mathematically is difficult, also because of the computational complexity. Data on actions and expectations, social characteristics, and institutional frames derived from empirical investigations are often not sufficiently separable as needed in order to properly analyze social interactions.
Having said this, Ioannides should be praised to have written a stimulating book that tries to interrelate social and spatial levels of complex economic phenomena. It is worth noting that this is something that should have more followers also in econometrics.
Return to Contents
of this issue