"Why the mortgage crisis could be a €40 bn problem"

Brendan Burgess

Founder
Messages
51,917
This is an example of how bad the analysis of our problem can be

Why the mortgage crisis could be a €40 bn problem

This is Dan White in yesterday’s Sindo with a picture of Morgan Kelly to add gravitas to the piece.

“it's hard to see how the Irish-owned banks can escape with losses of much less than €40bn on their mortgage books.”


Here is a summary of his figures
total mortgages||€100bn
Default rate|25%| €25bn
Loss on €60bn trackers|22%|€13bn
Total losses ||€38 b
This article is so full of mistakes, it’s just incredible

Total owner occupied mortgages of Irish banks: €74 billion
Total owned if you exclude 85% of Bank of Ireland: € 56 billion - not €100 bn

So he starts with almost double the relevant number. He never mentions residential investment mortgages, so it’s reasonable to assume that he is referring to owner occupied homes only.

“half of all Irish mortgages are under water …
House prices have further to fall…
The average house price is still 6 times average earnings…
Means that banks are looking at 25pc write down on their mortgage books”

It means nothing of the sort.

7% of the market is in arrears over 90 days. It will probably increase and it is higher for those in big negative equity. So let’s say that it increases to 20% and that they all default. And let’s say that the lender loses 60% on all these loans.

Total loans| €56 billion
In negative equity| €28 billion
20% default| €5.6 billion
60% loss on these| €3.36 billion
Total provided for in stress tests: €5.7 billion or €4.2 billion if you exclude Bank of Ireland.

How could you get a 25% loss on the mortgage book?
50% of mortgages won't be in negative equity, so the banks won’t lose on these.
You would have to assume that all mortgages in negative equity default
You get a 50% negative equity with a 100% mortgage and a price fall of 50%. Houses have fallen 50% from the peak, but most houses were bought outside the peak so the price falls have been lower. Some mortgages were 100% but the vast majority were not.
Some mortgages of less than 100% LTV will end up in over 50% negative equity due to arrears.

But it is almost impossible to come up with a 25% loss rate.

Even Morgan Kelly comes up with a 10% loss rate which is 20% of mortgages default with an average loss of 50%. (his estimate of €5bn to €6bn was based on 10% of €56 bn)

So how much will the banks lose on trackers
White says: 22% of €60 bn - €13 billion
Trackers form 50% of the €56 billion
Correct tracker figure: €28 billion

It’s very hard to calculate the loss on these. Certainly a third party would not pay €28 billion for these, when they could lend this money out at standard variable rate.
If you take into account the losses on trackers, then you should also take into account the future profit on non-trackers. They will balance each other out.
Even if you reject that argument, then you must accept that the defaults on trackers will be a lot lower.
If I have a mortgage of €200k and a house worth €100k, the maximum loss is €100k.
White adds the default rate to the 22% discount on trackers.
If the tracker has lost the bank 22% of €200k – that is €44k, so the additional loss is €54k or 54%
Many trackers will trade up and so surrender their trackers early. So the correct future losses on trackers is about 10% or €2.8 billion.
As of now, the banks are borrowing from the ECB at 1.5%, so it could be argued that they are making money on these trackers.

Summary of losses and assumptions
default losses|€3.36 bn
Tracker losses|€2.8 bn
Less double counted |1.5 bn
Net losses|€5 bn
Although only 7% of mortgages are in arrears and most will recover, it is assumed that 20% will actually default through voluntary surrender or repossession.
The bank will recover only 40% of the value of defaults.
Trackers should be written down by 10%
 
But facts don't make good copy in papers when they are trying to make a story sound worse.
 
Back
Top