Should the Central Bank's Loan to Income limits be scrapped?

[Brendan Burgess]

The LTI limits are a very crude measure.

I would not lend 3.5 times the income to someone on €20,000.

I would consider 4 times the income to someone on €100k as they would have far more disposable income.

I would lend a much higher multiple to couple of 25 year old newly qualified accountants than I would to a 35 year old bus driver as I would expect their income to rise faster

So the right solution is the net affordability. But that would be impossible for the Central Bank to codify.

So let the lenders operate their own limits on income.

In any event, there is a strong correlation between Loan to Income and Loan to Value.

Brendan

 

[Brendan Burgess]

The Central Bank has some research on this

Macro-prudential Tools and Credit Risk of Property Lending at Irish banks

If borrowers with the highest LTV also extend themselves furthest in terms of the LTI, then an either or approach will remove the highest risk loans from the book. In the Irish case, Figure 6 shows there is a strong positive relationship for low to moderate LTV and LTI levels for the whole sample. This relationship breaks down for originating LTVs between 80 and 90 per cent, followed by a strong relationship for the very highest LTV loans issued. Given the large portion of lending in the 80 to 90 per cent LTV region, a combination of both tools is required to reduce risk on the repayment and collateral channels.

 

[odyssey06]

That a Central Bank in 2016, with all the resources at its disposal in terms of IT and financial information, is relying on as a crude a measure as an LTI ratio is an absolute joke. It's the level I'd expect from someone doing a college thesis.

 

[Brendan Burgess]

This relationship breaks down for originating LTVs between 80 and 90 per cent,

I don't understand this point. Can anyone else explain it?

If I read the graph correctly, at 79.4% LTV, the LTI is 3.4 times.

At 90.5% LTV, the LTI is about the same.

The overall message of the graph seems to be that when people get high LTV loans, the LTI is also very high. Reckless lending is reckless lending.

 

[Sarenco]

Ronan Lyons made a very strong case, in my opinion, that the Central Bank's macro prudential rules should be restricted to LTV limits:-

[broken link removed]

 

[trasneoir]

The LTI limits are a very crude measure.

Fair enough. But given the current public appetite, I can't see us agreeing to a more "sophisticated" system that isn't also much more "lenient".

I would not lend 3.5 times the income to someone on €20,000.

And the central bank won't force you to. Banks are free to implement more sophisticated/stringent affordability criteria above the minimum limit set by the CB.

I would consider 4 times the income to someone on €100k as they would have far more disposable income.

Banks and customers have proven to be unreliable at assessing the risk of the customer's circumstances changing for the worse. If I lose my 100k job at some point in the next 20/30 years, then the balance of a 350k mortgage may be significantly more affordable than a 400k one. If the loan ends in default, then the cost to the bank's shareholders (or the state) for the smaller loan is also that much smaller.

In any event, there is a strong correlation between Loan to Income and Loan to Value.

In cases where LTI and LTV happen to be well aligned, then people will jump through both hoops at the same time, and that's fine. That doesn't mean that both aren't useful. If I make 20k per year, I won't be able to afford a 100k mortgage, regardless of whether I have plowed another 100k of inherited money into equity.

 

[Gordon Gekko]

I would contend that in an armageddon scenario (e.g. job-loss or illness), whether your mortgage is €400k or €350k would be a moot point.

In my own case, I'd be bunched regardless of whether my mortgage was +/- 20%.

 

[cremeegg]

This relationship breaks down for originating LTVs between 80 and 90 per cent.

I don't understand this point. Can anyone else explain it?

You do love a bit of data Brendan. The answer to your question is that the blue line is flat between 80% and 90%.

I would like to add a question of my own, what lambda has been used to compile the spline?

From just looking at the graph it would appear to be too high. I say this because the number of data points below the curve of the spline in the <20% range seem very much higher than the number of data points above the curve.

 

[Sarenco]

In my own case, I'd be bunched regardless of whether my mortgage was +/- 20%.

Bear in mind that the rules are designed to prevent banks from lending at imprudent levels - they are not designed to protect individual borrowers.

I agree with Ronan Lyons that there is no theoretical justification for the LTI limits. However, in practice, this seems to be largely an academic issue. Judging by the Central Bank's loan level data, the average home loan borrower (both FTB and SSB) falls well below the 3.5 LTI limit in any event.

I think the Central Bank should just drop the LTI limits and leave this to the loan underwriters - there are just too many variables at play for any crude LTI limit to be meaningful.

I absolutely agree with your point regarding the relevance of the absolute size of the loan, which is why I think it would make sense to move to a tiered approach for all borrowers (FTBs and SSBs) and to introduce a new tier for "jumbo" mortgages.

My suggested tiers would look like this:-

• €0 to €250k - 90% LTV
• €250k to €500k - 80% LTV
• €500k + - 70% LTV

I would also remove the ability of the banks to make exceptions from the LTV limits (save perhaps for borrowers in NE) and prohibit banks from offering cash-back incentives on the basis that they can be used to partly circumvent the LTV limits.

 

[Gordon Gekko]

Hi Sarenco

Should it not be the other way around?

i.e. higher LTV for bigger mortgages?

 

[Sarenco]

Should it not be the other way around?

I don't think so.

The whole point of the macro prudential rules is to limit losses to banks where borrowers default. Bigger mortgage = bigger potential loss and therefore a stricter underwriting standard is appropriate.

This basic principle is already reflected in the higher LTV that applies to FTBs borrowing over €220k - I'm just extending this to all borrowers (FTBs and SSBs) and introducing a third tier for jumbo mortgages.