"Location efficiency" and the probability of mortgage default

Iwannahurl

I'm not sure whether this is the most appropriate forum for this topic, since it touches on a number of inter-related themes including mortgage lending practices, land use and transportation planning.

This interesting American study is in a new (and rather curious) journal called the Journal of Sustainable Real Estate: Location Efficiency and Mortgage Default.

Based on a sample of over 40,000 mortgages in the metropolitan areas of Chicago, Jacksonville and San Francisco, the study found that the probability of mortgage default increases with the number of vehicles owned, and decreases with higher Walk Scores (ie 'walkability' or absence of car dependence) in high income areas.

According to the authors (two of them professors from the universities of Alabama and Florida respectively) the results provide justification for "smart growth development and urban revitalization" policies, because designing neighbourhoods that reduce motor vehicle ownership and use is beneficial to borrowers and banks as well as the environment.

Whatever one might think of the "smart" jargon, I wonder whether there might be any lessons for Ireland? If Morgan Kelly is right, there are a lot more mortgage defaults on the way. Will the patterns and predictors of Irish mortgage defaults be unique to our situation, and if so what can we learn re future mortgage lending as it relates to housing policy?

AlekSmart

Because designing neighbourhoods that reduce motor vehicle ownership and use is beneficial to borrowers and banks as well as the environment.

But Iwannahurl,it`s fluck all use to a Minister for Transport who has signed a Top-Secret "Commercially Sensitive" agreement with Toll-Facility operators guaranteeing them a set-level of income over the life of the contract.

Jeepers man...do you think Noel Dempsey would do manything other than burn copies of this report on the steps of Navan Courthouse....the more ribbon developments and ghost estates this cabal of creepy crooks can keep going the better their retirement pot will be.....QED ! :eek:

later12

That's an interesting study OP, but simply in terms of risk analysis I wonder if it is more relevant to US financial institutions. The US mortgage holder typically has a more obvious aversion to public transport than the Irish or the European equivalent might have.
That is significant in light of the weight that the report's authors contribute to automobile and fuel costs in acting as an economic buffer.
It is also significant assuming, as i'm sure the authors assume, that if a family have less need for a car, they will use the money they would have so spent, wisely - either by adding to their savings or via mortgage repayments. That is not necessarily the case in the Republic of Ireland, knowing what we do about Irish spending patterns. I can't help but feel that if the same Irish families had no need for cars they would have bought them regardless, or else blown the money on plasma TVs or other discretionary spending.

swampgas

Iwannahurl wrote: »

I'm not sure whether this is the most appropriate forum for this topic, since it touches on a number of inter-related themes including mortgage lending practices, land use and transportation planning.

This interesting American study is in a new (and rather curious) journal called the Journal of Sustainable Real Estate: Location Efficiency and Mortgage Default.

Based on a sample of over 40,000 mortgages in the metropolitan areas of Chicago, Jacksonville and San Francisco, the study found that the probability of mortgage default increases with the number of vehicles owned, and decreases with higher Walk Scores (ie 'walkability' or absence of car dependence) in high income areas.

According to the authors (two of them professors from the universities of Alabama and Florida respectively) the results provide justification for "smart growth development and urban revitalization" policies, because designing neighbourhoods that reduce motor vehicle ownership and use is beneficial to borrowers and banks as well as the environment.

Whatever one might think of the "smart" jargon, I wonder whether there might be any lessons for Ireland? If Morgan Kelly is right, there are a lot more mortgage defaults on the way. Will the patterns and predictors of Irish mortgage defaults be unique to our situation, and if so what can we learn re future mortgage lending as it relates to housing policy?

Interesting read.

It certainly makes some of those housing estates built during the bubble look even less attractive both to potential buyers and to potential mortgage lenders.

And it indicates another downside to the way housing in Ireland has been built up as a low-density suburban sprawl, which generally requires one or two cars per household.

GSF

I'm not sure I understand what the correlation is here between car owning & defaulting. Is it the cost of owning & running a car? Presumably it cant be that simple.

Is it more than new suburbs are furtherest from town centres and these were bought by mainly young new homeowners who still have not paid off much of their mortgage?

later12

GSF wrote: »

I'm not sure I understand what the correlation is here between car owning & defaulting. Is it the cost of owning & running a car?

Partially, yes.

There is also the assertion that location-efficient homes are more likely to hold their value, although again the magnitude of relevance to Ireland is, from a risk analysis viewpoint, ambiguous since foreclosures are a different business entirely in the US.

Another factor which does not readily apply to Ireland is 'walk score' - that those in low income high default neighbourhoods tend to be in urban areas, while those in low default high income areas tend to be further away from the urban centre. In Ireland, that does not really apply for various reasons.

This is very much an American study, and while interesting, cannot really take account of the situation here nor have its conclusions translated directly onto the Irish property model.

Scofflaw

GSF wrote: »

I'm not sure I understand what the correlation is here between car owning & defaulting. Is it the cost of owning & running a car? Presumably it cant be that simple.

Is it more than new suburbs are furtherest from town centres and these were bought by mainly young new homeowners who still have not paid off much of their mortgage?

Essentially, the suggestion is that it is pretty much that simple - that because the second major cost in US households is transportation, 'location-efficient' homes (that is, close to amenities such as shops and schools), by allowing a reduction in that cost, are at less risk of default because they have an additional buffer, all other things being equal. In other words, if you look at default risk for a given income/mortgage ratio, the default risk is lower for a mortgage holder who need not drive to do anything, because less of their income needs to be taken up by transportation costs.

I don't see that such a conclusion doesn't translate to Ireland, since it would apply anywhere that transportation is a major household cost factor.

So, the only question is whether transportation - in particular, car ownership - is a major household cost factor in Ireland. Looking at this thread, where people estimate their car costs at 190/month to 750/month, it appears likely that car ownership costs are probably the single largest 'discretionary' household cost after mortgage costs.

In turn, that means the study's essential conclusions are applicable in Ireland. The US-specific parts, such as how they calculate Walk Scores, are presumably not as easily transferable, but the essential conclusion that because running a car costs money, buying a house in an area where you don't need a car gives you an additional financial buffer against default seems completely and obviously applicable anywhere.

cordially,
Scofflaw

ei.sdraob

How does this study apply to Ireland?

They have bankruptcy laws where its easy to walk away, we dont

Apples and oranges

herya

that, and also if you buy in the area where you are forced to commute loads, it's probably not by choice, but because that's all you could afford, therefore your finances are less solid to start with and more prone to mortgage issues should your situation change even slightly.

Permabear

This post has been deleted.

later12

Scofflaw wrote: »

but the essential conclusion that because running a car costs money, buying a house in an area where you don't need a car gives you an additional financial buffer against default seems completely and obviously applicable anywhere.

I wouldn't say it is so universally acceptable. In many cities, such as Dublin, if you live in an area where you don't need a car you are likely to be paying significantly more for your mortgage to begin with, and that increased cost is not always likely to be met by reverting to public transport.

That is not necessarily so in the US where their road infrastructure and heavy reliance on cars typically alleviates undesirability of what the equivalent in Ireland might be seen as a more undesirable satellite location - for example Swords or Cabinteely, compared with somewhere like Harold's Cross or Ranelagh.

The latter are typical high income locations with high walk value and high house prices, the former are typical lower income satellite locations with low walk value, with lower house prices. They are typical of the trend here.

Also, although fuel costs are significant right now, they have more relevance in rural locations. If a family are going to default on their mortgage in Dublin, they are probably going to default whether they have a car or not. I don't think anybody would realistically default on their mortgage to keep their car, basically. Again, that could be different in the US.

Scofflaw

Goodness me - enough red herrings to start a fish market!

How does this study apply to Ireland?

They have bankruptcy laws where its easy to walk away, we dont

Apples and oranges

The differences in bankruptcy laws are completely irrelevant, since what's essentially being described here is difficulty in paying the mortgage. If in Ireland that difficulty translates to arrears rather than default, that's how it translates.

that, and also if you buy in the area where you are forced to commute loads, it's probably not by choice, but because that's all you could afford, therefore your finances are less solid to start with and more prone to mortgage issues should your situation change even slightly.

As long as you've bought a house at the same multiple of income as someone who has bought a house in a more desirable area, the effect described still obtains.

I wouldn't say it is so universally acceptable. In many cities, such as Dublin, if you live in an area where you don't need a car you are likely to be paying significantly more for your mortgage to begin with, and that increased cost is not always likely to be met by reverting to public transport.

That is not necessarily so in the US where their road infrastructure and heavy reliance on cars typically alleviates undesirability of what the equivalent in Ireland might be seen as a more undesirable satellite location - for example Swords or Cabinteely, compared with somewhere like Harold's Cross or Ranelagh.

The latter are typical high income locations with high walk value and high house prices, the former are typical lower income satellite locations with low walk value, with lower house prices. They are typical of the trend here.

Again, it's completely irrelevant what the absolute prices are.

Also, although fuel costs are significant right now, they have more relevance in rural locations. If a family are going to default on their mortgage in Dublin, they are probably going to default whether they have a car or not. I don't think anybody would realistically default on their mortgage to keep their car, basically. Again, that could be different in the US.

That's why I pointed to the thread where people calculated their monthly costs of owning and running a car - to show that those costs are significant. And the point is that there are locations where you effectively have to default (or go into arrears) to keep your car, because there is no other option.

Let's go through a worked example, and see if that helps people.

Couple 1: 4-bed house in Swords estate - price €400k, loan €368k, variable rate mortgage repayments €1,489. Couples income 2 x €33k (€60k after tax). Car costs €500/month first car, €320/month second. Both work city centre. Nearest local amenities 3 miles by road.

Couple 2: 2-bed house in Ranelagh street - price €400k, loan €368k, variable rate mortgage repayments €1,489. Couples income 2 x €33k (€60k after tax). Car costs €500/month first car, €320/month second. Both work city centre. Nearest local amenities 700 yards.

OK, so all the ceteri paribus are set equal - the only differences are the size of house in question, and the commute/walk distances.

The question the study addresses is a simple one - of those two couples, which is more likely to have difficulties paying their mortgage? The answer, according to the study, is Couple 1, because they have no options bar their cars, whereas Couple 2 could easily downsize their cars and still be able to commute and shop. They therefore have a financial buffer that Couple 1 do not have.

Hardly complicated, surely?

cordially,
Scofflaw

later12

Scofflaw wrote: »

The differences in bankruptcy laws are completely irrelevant, since what's essentially being described here is difficulty in paying the mortgage. If in Ireland that difficulty translates to arrears rather than default, that's how it translates.

The question the study addresses is a simple one - of those two couples, which is more likely to have difficulties paying their mortgage?

Difficulty indicators are one thing, the OP's question was about predictors of mortgage defaults - and this study, unfortunately, is not really valuable or translatable in that sense, for Ireland.

Let's go through a worked example, and see if that helps people.

Couple 1: 4-bed house in Swords estate - price €400k, loan €368k, variable rate mortgage repayments €1,489. Couples income 2 x €33k (€60k after tax). Car costs €500/month first car, €320/month second. Both work city centre. Nearest local amenities 3 miles by road.

Couple 2: 2-bed house in Ranelagh street - price €400k, loan €368k, variable rate mortgage repayments €1,489. Couples income 2 x €33k (€60k after tax). Car costs €500/month first car, €320/month second. Both work city centre. Nearest local amenities 700 yards.

OK, so all the ceteri paribus are set equal - the only differences are the size of house in question, and the commute/walk distances.

Given the differences in Walk Value, why are the motor costs equal in your scenario? Surely walk values and fuel costs are negatively correlated.

Again, we have a vastly different public transport situation than exists in the US which tends to be (even) more car dependent than we are. That study uses vehicle normalised score and a walk score as a measure of location efficiency - no public transport efficiency/ takeup score.

The question the study addresses is a simple one - of those two couples, which is more likely to have difficulties paying their mortgage? The answer, according to the study, is Couple 1, because they have no options bar their cars, whereas Couple 2 could easily downsize their cars and still be able to commute and shop. They therefore have a financial buffer that Couple 1 do not have.

Hardly complicated, surely?

Hardly so simplistic, surely. The study generally goes on the assumption that higher walkability means lower transport costs and that the greater disposable income is the liquid buffer.

In theory this would indicate that Couple 2 have a greater buffer, but precariously assumes that, rather against the trend, Couple 2 did not use their greater disposable income to otherwise invest in a non liquid non motor asset... like a conservatory or whatever.

silverharp

it seems pretty obvious , the majority of new properties were built in outer Dublin, D4 D6 etc had very little new capacity built apart from apartments. Mortgage defaults are going to be concentrated in anything built in the last 7 years or so. Most people living in the "nice" mature parts of Dublin were bought pre 2000 or traded up with equity from a previous property so will be sitting pretty. At least for Ireland its more a demographic issue then a transportation issue.

Permabear

This post has been deleted.

Scofflaw

later10 wrote: »

Difficulty indicators are one thing, the OP's question was about predictors of mortgage defaults - and this study, unfortunately, is not really valuable or translatable in that sense, for Ireland.

Um, yes, if one attempted to apply it without any thought whatsoever. The reason for default in the study is difficulty in paying the mortgage, and the difficulty in paying the mortgage translates across from the US. In Ireland, it will not give the same incidence of default because of the different legal regime here, but the relation between difficulties in paying and transport costs will still exist.

Default incidence (R) = difficulty in paying (P) / difficulty of defaulting (D).

For Ireland, D might be twice or more what it is in the US. So the same difficulty in paying (P) will result in a different default rate in Ireland and the US:

R eire = P / D eire
R US= P / D US

So for the same difficulty of payment P - say we set P equal to 3 - the Irish default rate will be the same fraction of the US default rate as US default difficulty is of Irish default difficulty. Say D eire is three times D US.

R eire = 3 / 3 = 1
R US= 3 / 1 = 3

So for any given level of difficulty in paying a mortgage, three times as many US mortgage holders will default as Irish mortgage holders.

But - and this is the important point - if you double P, the difficulty in paying, you double the rate of default, whether you're talking about Ireland or the US:

R eire = 6 / 3 = 2
R US= 6 / 1 = 6

Since the study establishes a correlation between transportation costs and difficulty in paying, the study shows that increased transportation costs at a given mortgage debt/income ratio increase the likelihood of default - a conclusion that applies no matter what the initial rate of default.

I admit that I'm a little alarmed at the number of people who can't see the principle for the details here.

cordially,
Scofflaw

later12

Scofflaw wrote: »

I admit that I'm a little alarmed at the number of people who can't see the principle for the details here.

But the problem is not necessarily the principle - it's primarily the study itself.

Yes of course one could design a location efficiency study for Irish mortgage holders - but the one cited by the OP is, although interesting, not it.

Firstly, however, I have some serious reservations about your post above.

Presumably one would theoretically measure the difficulty in or the resistance to legal default by measuring the incidence of default cases lodged in the Chancery or orders instituted with respect to the volume of mortgages in arrears or the volume of seriously criticised mortgages (or any other marker) which would otherwise be in legal process according to some international calibration.

Unfortunately this is not so simplistic as you have apparently imagined.

What about lender effect? Between government owned, government majority and other private lenders, the Irish mortgage market is quite a heterogenous and ever changing one; and unlike the American situation, house price and lending data is not nearly as transparent.
And what of the very real, and I would suggest, even more significant issue of default stigma. This present phenomenon is in many cases, I would suggest far more significant than the motor status of a mortgage lender, yet would be extraordinarily difficult to accurately quantify in this fast moving debt environment.

There are many other concerns I would have, but not to harp on about it, I really do think that public transport is another challenging issue. One would have to measure public transport take-up, availability, and possibly efficiency. I would suggest the public transport issue would be far more relevant than car ownership, at least in urban centres.

So yes. One could indeed create a location efficiency study for Ireland. But by the time you'd be finished with it, it really probably wouldn't resemble this study at all and that is why I am simply saying that this particular US study is untranslatable to the Irish situation.

cyclopath2001

Iwannahurl wrote: »

Based on a sample of over 40,000 mortgages in the metropolitan areas of Chicago, Jacksonville and San Francisco, the study found that the probability of mortgage default increases with the number of vehicles owned, and decreases with higher Walk Scores (ie 'walkability' or absence of car dependence) in high income areas.

There could be indirect advantages to a less car-oriented lifestyle, such as health advantages from more walking or cycling. Better health would improve the people's ability to retain and perform at work.

Additionally, if the person lives in an area with higher density of employment opportunities (e.g. Dublin) and they can reach them on foot, cycling or by public transport, they have better employment mobility than someone living outside a rural town with one or two major employers, needing to drive for anything at all.

This all points to the advantages of city living and of concentrating employment around cities where the transport, education and health needs can be efficiently provided.

cavedave

There is more research on the UK than Ireland but we have a similar social and legal framework (though I think our debt loans are significantly harsher)

This paper on the UK gives some time series analysis methods for predicting defaults. It used mainly arrears and loan-to-value ratio
macroeconomic variables seem to be of less importance.

Another paper on UK default rates suggests investment properties (which are particularly popular in Ireland) are risky
'Fitch believes a borrower is even more likely to default on an investment
property than on a second home. Accordingly, Fitch increases base default rates by 10%-35%.'

If LTV and Arrears are the main predictors of mortgage default in the UK with whether the property is owner occupied of some importance. Macro economic factors like GDP growth rate, unemployment, tax rates and even legal framework are believed to be of less importance. I think this is because many of these significantly correlate with LTV and arrears.

As Morgan Kelly pointed out Irish banks don't want to kick people out of their homes because that reduces the value of their balance sheet but the guys who the Irish banks owe money to have no such qualms.

Does anyone have a historical dataset of arrears, LTV etc figures? It may be easier to just run some regressions and let the figures do the talking.