Moving to opportunity

Recent calls by politicians for strategies to ‘revitalise flagging towns’ and ‘tackle areas of poverty’ have brought new attention to the effect that neighbourhoods have on the socio-economic outcomes of their residents. Prior to the last election, the Government’s Regional Australia Summit and the Opposition’s call for Education Priority Zones indicated that both parties were paying more attention to geographical patterns of disadvantage than at any time in the past 20 years. The assumption that locational disadvantage breeds personal disadvantage appears to enjoy unusual cross-party support. But intriguingly, social science has provided very little justification for this commonsense notion.

Until now, that is. In the US, a radical social experiment is starting to provide some dramatic insights into so-called "neighborhood effects". Initiated in 1994, the Moving to Opportunity program provides rental subsidies to low-income families in poor neighbourhoods, so long as these funds are spent on moving to a low-poverty area.

This program is also a social experiment in the truest sense of the term, since those who were selected to receive housing vouchers were selected by lottery. Data collected prior to the lottery revealed extreme levels of disadvantage in the very worst areas, with health problems, violence, drugs and poverty prevalent in some neighborhoods. Whether Australian politicians from leafy suburban electorates have much interest in life in our own pockets of poverty remains an open question.

But in understanding locational disadvantage, it is not enough to compare outcomes in different areas. The problem is that we do not know whether the association between personal disadvantage and neighbourhood disadvantage simply reflects the fact that the poor can only afford to live in poor neighborhoods. Or as one of the leading researchers on neighborhood effects, Harvard Professor Jeffrey Liebman, puts it: "academic researchers had generally failed to find evidence of large neighborhood effects, largely because it was difficult to find truly comparable individuals living in different neighborhoods." MTO helps answer that question, by following the life outcomes of lottery winners and comparing them with lottery losers. And according to Liebman, this "randomized design establishes quite clearly that the neighborhood in which a person resides affects important outcomes."

Those who won the MTO lottery and moved to a low-poverty area reported substantially lower levels of exposure to violence - while those families who stayed in poor neighborhoods reported continuing high levels of fear. Moving had benefits in a range of other dimensions, too. Mothers who moved reported being healthier, feeling calmer and less prone to episodes of depression. For younger children, moving boosted test scores for both reading and mathematics. Among older children, moving reduced absenteeism, lowered school dropout rates, and – at least among boys – lessened behaviour problems. Child health also improved, with asthma attacks declining markedly. The only mixed message comes from data on delinquency: while children who moved were less likely to be arrested for violent crimes, they were more likely to be arrested for property crime. However, it seems possible that this may reflect different police behaviour in the different environments rather than a true rise in delinquency.

What about welfare usage? Here, studying the effect of the MTO program was made more difficult by the sharp overall fall in welfare usage throughout the US during the 1990s. Yet researchers found some evidence that those who moved had somewhat higher employment rates and lower welfare usage than those who stayed.

Clearly, there are differences in the way poverty manifests itself in Australia and the US. While US poverty is concentrated in the inner city, the poor in Australia tend to live in outer suburbs. But nonetheless disadvantage in Australia is geographically concentrated, and increasingly so. While it seems plausible that a similar program may have dramatic effects in the Australian context, only a local trial can tell us for sure.

More importantly, a local MTO-style experiment may build the political capital needed for such reforms. The US program shifts the disadvantaged from poor areas into middle America. With the ability of NIMBY politics to rapidly change the electoral landscape, it remains to be seen how marginal middle-class electorates will feel about an influx from poorer suburbs. Right now, both sides of politics propose sending money to disadvantaged areas, rather than running the risk of middle-class anger over being asked to share their neighborhoods with those who are less well off. If the Howard Government’s regional policy over the next three years is to become more than "whiteboard politics", it will need to move beyond shoring up marginal seats to really fixing social problems.