Is It Safe to Live in Areas with a High Presence of Social Housing?

Reading Time: 5 minutes

Introduction

A Google search about safety in Dublin would result in all sorts of horror stories about dangerous neighbourhoods to be avoided. Many of the claims therein are based on personal experiences and opinions. In this analysis we have a close look at crime rate in Dublin using Census data and crime data in the Irish Central Statistics Office [CSO]. 

Additionally, we attempt to shed lights into the relationship between social housing and crime. This is motivated by a hypothesis that areas with a high presence of social housing tend to have symptoms of social and economic vulnerability. In recent years there has been a lot of focus on building social housing in Ireland. At the country level, new built social housing has increased 8 times between 2016-2019.  Looking ahead, there are numerous government schemes to build a large number of social housing units in coming years. With this trend, there are likely to be more negative views toward the “dangerous” referenced in the first paragraph of this blog.

At the time of writing this blog, Daft reports that housing prices have risen 7.6% on Q1 2021. Consequently, house hunters might need to find more affordable options such as moving into areas that are not recommended on social media. This prompts a sensitive question: is it safe to live in a neighbourhood with a high presence of social housing? We are going to use a data-driven approach to flesh out a detailed answer to this hypothesis-driven question.

The Data Source

Social Housing data is available in Theme 5, Private Household Census. The data comprises the number of households provided by local authorities and other housing bodies in defined areas. Crime records by Garda station can be pulled from the CSO’s crime database. We approximately aggregate the raw social housing data by areas defined within each Garda boundary. Thus, for a given area we have data on total households, social housing units, population, and reported crime. We focus on 2016 as a snapshot in time for both social housing and crime data; 2016 is the year whence complete Census and crime data are both available (although it is unlikely that either Census or crime rate has changed drastically over the last few years).

Pattern of Social Housing in Dublin

Figure 1 shows that Tallaght is the area with the highest number of social housing units. After scaling by the total households within the respective area to achieve percentage measures of social housing, Ballymun takes the pole position. Pattern distribution of social housing in Dublin is reported elsewhere and we do not replicate the details here; See for example for an interesting discussion on the contrast division between Dublin City Centre, the West, and the East, and the rest of Dublin.

Top 5 Crime Spots in Dublin

Time series of total crime recorded in Dublin over the last 18 years is depicted in Figure 2. The crime trend is relatively consistent in time for the most part of the last 18 years, with an exception in the strong reduction from 2019 to 2020 most likely due to COVID19. At an overall crime level, the top 5 crime hotspots in Dublin: Pearse Street, Store Street, Bridgewell, Tallaght, and Blanchardstown.

On average these areas experience between 3 to 6 times higher risk of crime than the baseline; We define the baseline as the rest of Dublin’s average. Using the same logic, top 5 areas for burglary related crimes are Tallaght, Pearse Street, Blanchardstown, Dundrum, Rathfarnham having between 2 to 3 times higher risk than the rest of Dublin (i.e. the baseline for burglary crime). For drug related offences, Store Street, Tallaght, Pearse Street, Cooklock, Finglas are 3-4 times more risky. For anti-social offences, Pearse Street, Store Street, Tallaght, Kevin Street, Blanchardstown are 2-11 times more risky. The most extreme data point is Bridgewell: 83 times higher risk than the associated baseline with respective to organised crime.

Is There a Relationship Between Social Housing and Crime?

We first compute the population for the catchment areas covered by each Garda station to give an indication of crime rates per 10,000 head population. This is to ensure a fair comparison between areas with varying population density. See the population aggregation calculation report.

We then overlay the rescaled crime data with social housing percentage in ascending order as depicted in Figure 3. In comparison to unscaled data as discussed in Section IV, the high ranking (with respect to crime) in areas with large populations such as Tallaght, Blanchardstown, Finglas, Coolock has been replaced by areas around the city centre (Fitzgibbon Street, Kevin Street, Mountjoy). This implies that areas with an overall high-crime rate might have a low level of crime in certain sub-areas. Vice versa, the areas having low crime rate would consist of unpleasant sub-areas.

Exploratory data analysis reveals that there is not enough statistically significant evidence to conclude that there is a linear relationship between social housing and crime. That is,  a high presence of social housing doesn’t necessarily mean that the area has a high level of crime, and vice versa. We are aware of confounding factors such as housing quality, income level, tourism, etc. that would otherwise influence this relationship. Therefore we do not perform causal inference; Instead we objectively conduct exploratory data analyses and call out interesting patterns from the data.

In summary: our analysis demonstrates that, in the grand scheme of things, social housing and crime do not go hand in hand. On the one hand, Dublin has many neighbourhoods with a high presence of social housing yet considerably pleasant to live, i.e. recommended places to live and also with higher likelihood of price increase and positive return on investment. On the other hand, an area with little to no presence of social housing does not guarantee complete safety. Renters, buyers, investors beware of exceptionally dodgy neighbourhoods in Dublin that are ravaged by criminal activities, i.e. places to avoid living.

As we deep-dive into social housing and crime type at a lower areal level, as per Figure 4, we find that burglary risk appears to be associated with the likelihood of wealthy areas/good homes (Dun Laoghaire, Cabinteely, Blackrock, Dundrum) with some exceptions that are in the city centre. Alas, this is perhaps not a surprise finding. Per 10,000 head population, the areas with low burglary risk in Dublin include: Lusk, Raheny, Cabra. 

Summary

• There are staggering differences in crime rates across Dublin. 
• There is not enough evidence to link social housing with crime risk. There are areas with a high proportion of social housing yet with little or no presence of crime. In areas with an overall high crime level, there exist sub-areas with absence of crime and are definitely worth living.
• Considerably wealthy areas are targeted for burglary crime.

Final Words

If you’re serious about choosing a property for living or investing alike, you’d want to know about the safety of your future neighbourhood, and you’d want data and unbiased actionable insights. At WhereToLive.ie we are passionate about building a data-driven approach to a perfect home.

By all means, safety is not the only factor determining the value of your house or your neighbourhood. The price of the house, the transport infrastructure, public service availability, local school ranking, demographics of the area, etc. are all crucial for decision making with confidence. We’ll cover all of that in future blogs.