Do high prices deter fertility? Evidence from England and Wales
Washbrook, E. Do high prices deter fertility? Evidence from England and Wales.
This paper investigates the effect of the level of local area house prices on fertility in England and Wales in the period 1995 to 2008. Traditionally the relationship between the cost of housing and fertility has been assumed to be negative: high relative house prices deter the moves into home ownership or into larger dwellings deemed necessary by many for raising children. Here we test the hypothesis that this effect is weaker among home owners than among renters because of the wealth effects of increasing home values among owners. Individual-level data from a household panel are merged to area-level data on house prices. Effects are identified from spatial and temporal variation in house prices and fixed effects and individual-level control variables are used to deal with confounding. A key innovation is the use of monthly-level data combined with the estimation of a series of models varying the lag between house price measurement and the fertility outcome. This enables the exploration of the precise timing of the relationship between housing market conditions and fertility responses among different groups. The results reveal positive but temporary effects on the fertility of owners, in contrast to negative and longer-lasting effects on the fertility of renters.
This study focuses on the relationship between the housing market context and decisions about if and when to have children. Housing costs vary dramatically by area and over time, and theory suggests this variation is likely to affect fertility behaviour. The traditional view in demography is that high housing costs will tend to deter fertility. Children require space and appropriate family-friendly housing, which in the view of many means an owner-occupied home in a certain type of neighbourhood. When housing is more expensive (relative to other goods and services) it becomes more costly to fulfil these requirements and households may delay childbearing or have smaller families. However, a closer reading of the literature suggests that the impact of house prices on childbearing is likely to vary across sub-groups of the population and will be dependent on a number of contextual factors. In particular, house price inflation has a positive effect on the equity of home owners and increased liquid wealth via this mechanism could lead to increased demand for children and higher fertility rates among that group. Negative effects may then be expected to be concentrated among those in the rental sector who do not experience wealth effects to off-set the price effect.
In this study we model the probability that a woman conceives in a particular month as a function of average relative house prices in her local area. We use individual-level data on women aged 18 to 44 between 1995 and 2008 from the BHPS matched to Land Registry data on house prices at the local authority level for England and Wales. The key research questions are whether the effect of local house prices on the conception rate varies with the woman’s housing tenure and whether she already has children.
The longitudinal models employed allow us to account for a number of methodological issues that make it difficult to indentify the true effect of house prices on fertility. First, there is the problem that women with different fertility preferences will tend to sort into particular types of neighbourhood. For example, career-oriented women may choose to live in large cities, where house prices are high, for labour market reasons and because they have relatively strong tastes of urban amenities. If such women also tend to desire fewer children then we risk erroneously attributing lower fertility rates to high city housing costs. The inclusion of area fixed effects in a longitudinal fertility model provides a way of controlling for any unobserved influences on fertility associated with a particular location that do not change over time.
A second source of potential bias is the fact that temporal variation in house prices will be correlated with macro factors such as interest rates and overall economic activity, which themselves may have independent effects on fertility. The inclusion of year fixed effects in the model accounts for the confounding of house price variation with any trends such as these that occur at the national level. A control for the local monthly unemployment rate further protects against contamination from shifts in economic conditions that are specific to a particular region or area.
Third, it is likely that there is a lag in the effect of a change in house prices on the conception rate and this lag may vary for women at different stages of the life course. For example, high house prices may lead young women to delay leaving the parental home, moving in to home ownership or forming a partnership. The subsequent impact of these delayed life events on fertility outcomes may only become apparent after several years have passed. The crude temporal aggregation embodied in annual-level measures of house prices and fertility means this form of data is of limited use in investigating the timing of impacts in a precise manner, and important relationships may be missed. Here we use monthly-level data and estimate models that build a profile of the impact of a house price change at a particular point in time on conception rates over the subsequent 36-month period.
The results confirm the importance unpacking the aggregate relationship between house prices and fertility, as the average association disguises a mixture of positive and negative, and long-run and short-run effects. For home owners, we find evidence of positive but temporary effects of house prices on fertility. A 10% increase in the average house price in a given month increases the fertility of home owners by 2.4% in the following year. However, cumulative fertility rates return to normal again by the time two years have passed, suggesting the effect for home owners is on fertility timing rather than completed family size. Additional analyses show that the positive fertility effects seen for owners are restricted to second- and higher-order births. In contrast there is evidence of delayed negative effects for renters. A 10% increase in the average house price in a given month has no significant effect on the fertility of renters in the following 12 months. However, two years later, fertility rates are lower by the order of 2 to 4%. Fertility rates for this group remain lower three years later, suggesting house price increases affect completed fertility as well as the timing of births. Effects are slightly weaker for those who began the period living in social rented accommodation, relative to those in the parental home or private rented accommodation. The negative fertility effects seen for private renters, like those for home owners, are concentrated in second- and higher-order births.