Do Neighbourhoods Have an Influence on Reproductive Intentions? Empirical Evidence from Milan

Abstract

Meggiolaro S. Do neighbourhoods have an influence on reproductive intentions? Empirical evidence from Milan, Regional Studies. This paper aims at studying the contextual effect on reproductive intentions with an innovative approach. It considers women and the groups to which they belong ‘close up’, using data in very great territorial detail with reference to an Italian metropolitan area: Milan. This allows context to be described in an in-depth and original way, thus allowing an examination to be made of the mechanism underlying the formation of fertility intentions. Results suggest that besides the importance of individual factors and preferences, reproductive choices are also influenced by some characteristics of the places in which individuals live.

Meggiolaro S. Les voisinages, influencent-ils les intentions de reproduire? Des preuves empiriques provenant de Milan, Regional Studies. Employant une façon innovatrice, cet article cherche à étudier l’impact du milieu sur les intentions de reproduire. A partir des données régionales très détaillées sur une zone métropolitaine, à savoir Milan, on considère ‘en gros plan’ les femmes et les groupes auxquels elles appartiennent, Cela permet la présentation du mileu d’une façon approfondie et originale, facilitant un examen du mécanisme qui étaye l’établissement des intentions de reproduire. Les résultats laissent voir que le choix de reproduire est influencé aussi par certains caractéristiques du milieu où habitent l’individu, à part l’importance des facteurs et des préférences particuliers.

Intentions de reproduire Impact du milieu Données sous-municipales Milan

Meggiolaro S. Wirkt sich die Nachbarschaft auf die reproduktiven Absichten aus? Empirische Belege aus Mailand, Regional Studies. In diesem Beitrag wird der Kontexteffekt auf reproduktive Absichten mit Hilfe eines innovativen Ansatzes untersucht. In einer ‘Nahaufnahme’ untersuchen wir Frauen und die Gruppen, zu denen sie gehören, wofür äußerst detaillierte territoriale Daten des italienischen Metropolitangebiets Mailand zum Einsatz kommen. Auf diese Weise lässt sich der Kontext auf ausführliche und originelle Weise beschreiben, was wiederum eine Untersuchung der Mechanismen ermöglicht, die der Bildung von reproduktiven Absichten zugrundeliegen. Aus den Ergebnissen geht hervor, dass die Reproduktionsentscheidungen nicht nur von individuellen Faktoren und Vorlieben abhängen, sondern auch von einigen Merkmalen der Orte, an denen die Personen wohnen.

Reproduktive Absichten Kontexteffekte Details unterhalb der Gemeindeebene Mailand

Meggiolaro S. ¿Influyen los vecindarios en las intenciones reproductivas? Evidencia empírica de Milán, Regional Studies. La finalidad de este artículo es estudiar el efecto contextual en las intenciones reproductivas con un enfoque innovador. Analizamos en primer plano a las mujeres y los grupos a los que pertenecen usando datos con gran detalle territorial con referencia a un área metropolitana de Italia: Milán. De esta manera podemos describir un contexto de un modo profundo y original, lo que nos permite hacer un análisis de los mecanismos subyacentes en la formación de las intenciones de fertilidad. Los resultados indican que además de la importancia de factores y preferencias individuales, las decisiones reproductivas también están influenciadas por algunas características de los lugares en los que viven las personas.

Intenciones reproductivas Efectos contextuales Datos submunicipales Milano

Keywords:

• Reproductive intentions
• Contextual effects
• Sub-municipal detail
• Milan

JEL classifications:

• R
• R2
• R23

Acknowledgements

The author would like to thank Fausta Ongaro and Nicola Torelli for helpful and valuable comments on drafts of this paper; and Stan Panis for his useful suggestions on the use of the aML software and model specification. The author would also like to thank the Associate Editor and two anonymous referees for their comments that greatly improved this paper.

Notes

Concerning this remark, a previous study with reference to Milan showed some neighbourhood effects in the process of loneliness perception among the elderly population (Micheli and Rivellini, 2001).

Short-term (rather than long-term) reproductive intentions are probably more affected by context than by long-term intentions: the latter does represent more deep-rooted orientations and, consequently, may be less influenced by contextual aspects.

See note 1.

In Northern European countries, social-democratic welfare governments are more prone to reconciling female labour market participation and children with intervention to the advantage of families. Conversely, Mediterranean countries are characterized by conservative welfare (Esping-Andersen, 1999).

As regards the effect of resident employment status, there are various theories on how they may influence fertility at micro- and macro-levels, and hypotheses are not always univocal (for a review of the effects of women’s participation in the labour market, see Brewster and Rindfuss, 2000).

There is clearly a bi-univocal relation between child-bearing (plans) and marriage/divorce decisions (that is, there may be a causality problem). However, reproductive intentions are reached differently according to marital status, and must be standardized.

The reference to different years is not a limitation, as the social phenomena considered here change slowly over time.

The underlying hypothesis is that there is a link between the political–cultural climate of residence and the values systems of individuals. This hypothesis requires some caution: it presumes not only that there is an association between political orientations and the values system, but also that this association has consequences on the reproductive behaviour of individuals. In Italy, confirmation of this hypothesis is found at the macro-level, considering the link between referendum choices and reproductive behaviour (on the vote expressed for the 1974 abrogative referendum on divorce, see Mannheimer et al., 1978; also Livi Bacci, 1980; and Dalla Zuanna and Righi, 1999).

In particular, with aML software (Lillard and Panis, 2003), the multilevel residual structure is an explicit feature of the estimation procedure.

However, preliminary analyses considering models with random effects, estimated by means of an approximation of the likelihood given by the numerical integration of residuals and based on Gauss–Hermite quadrature (Abramowitz and Stegun, 1972), give similar results, with negligible variance of random components.

Never married women also include women in consensual unions, as in preliminary analyses they do not show different reproductive intentions with respect to never married women who do not live with a partner.

A close association is indeed observed between women who have already achieved high fertility (with three or more children) and women who go to church regularly. The variable parity, included in the model, distinguishes women without children, women with one child, and women with two or more children. Among the latter category are women with three (or more) children: most of them regularly attend church, but they have already achieved satisfactory fertility and so do not want other children. This may lead to the spurious effect observed in the model. However, it was not possible to distinguish women with two children from women with three or more children due to the small sample size.

aML software was used (Lillard and Panis, 2003).

In addition, the measures of association between predicted probabilities and observed responses, listed in the last row of Table 5, show how this model is improved with respect that of Table 4.

It is interesting to note that districts 12 and 15, where higher percentages of women who intend to have children in the short-term were found by the previous model (Table 5), are characterized by a low level of social unease and a high level of community vitality (for details, see Meggiolaro, 2005).