Comments on: More data on mortgage delinquency and downpayments A slice of lime in the soda Sun, 26 Oct 2014 19:05:02 +0000 hourly 1 By: MyLord Wed, 13 Jul 2011 17:23:43 +0000 This is a rather illegitimate view of the data as lending can never be safe once unsafe lending has inflated asset prices beyond all value. Allow unqualified lending and even qualified lending is a rather meaningless term. The real question is, if only qualified lending were allowed, how overvalued might housing have gotten and the answer is not much, but allowing unqualified lending can always raise prices above any margin of safety, even 20%.

By: KPARKPLACE Wed, 13 Jul 2011 15:58:24 +0000 What this really shows is a spike in defaults AT 80%. I bet you that loans with exactly 20% downpayment are lumped with 80-85% LTV and the 75-80% bucket only goes to 79%. That is just bad display of data.

Use the FHFA tables here: 3

and make your own buckets.

People stretched to qualify as conforming and had unusually high defaults precisely at 80% LTV. This would be expected for whatever they set for the QRM too.

By: TFF Wed, 13 Jul 2011 11:21:28 +0000 “Thus, it is not true that you cannot have a safe mortgage with a low downpayment – that is an incorrect conclusion.”

Those 0-3% downpayment loans had a 5% to 20% default rate each and every year. That might be lower than the 3%-5% downpayment loans (isn’t 3% the lower limit for an FHA loan?), but it isn’t what I would call “safe”.

I agree with your conclusion that “other factors are at play” in creating this discontinuity. But your final line reaches beyond your evidence.

By: publius5432 Wed, 13 Jul 2011 03:14:48 +0000 Let’s make this really simple, and take the extreme case of 2007.

In 2007, loans with 0-3% down had a default rate 1/4 that of loans with a 3-5% down. All else equal, that doesn’t make sense – why would someone with no skin in the game perform 4X better than someone with some skin in the game?

Obviously, some other factor was at work. Maybe no money down mortgages had a higher average FICO score than 3% down mortgages. Maybe they had lower average DTIs. More reserves. Etc.

The point is, something caused these mortgages to perform 4X better. Whatever that is, could be applied to other low downpayment mortgages to help them perform well.

Thus, it is not true that you cannot have a safe mortgage with a low downpayment – that is an incorrect conclusion.

By: TFF Wed, 13 Jul 2011 02:15:29 +0000 “This suggests that 20% is some “magic number”, but that doesn’t make logical sense – the results should be more continuous.”

Isn’t 20% the number at which you (typically) don’t need PMI? Those with substantial assets will free up enough to meet that 20% down, but may prefer to keep the rest invested.

By: Danny_Black Wed, 13 Jul 2011 02:11:19 +0000 publius5432, i suspect that is a mix of two things. The chancers with 0% down got washed out in 2006 and that there were stricter underwriting with nothing down vs a little bit down.

By: hsvkitty Tue, 12 Jul 2011 23:31:01 +0000 @publius5432 huh? Ok look at it it differently then. If you take the last 2 figures, which are 0-3 and 3-5 and add them together to be 5% ( which is the difference between the other percentages down as well) then you get over 50% default rate.

By: sanchk Tue, 12 Jul 2011 22:39:47 +0000 I applaud your ability to attract data and charts.

By: publius5432 Tue, 12 Jul 2011 22:31:00 +0000 Something seems wrong. Why the discontinuity between 20% down and 15% down, while 20-25% and 10-15% show a much smaller difference? That is, the data seems to jump between 20% and 15%. This suggests that 20% is some “magic number”, but that doesn’t make logical sense – the results should be more continuous.

Also, as the first poster noted, the people with no money down performed better than those with 3% down? That is totally illogical.

These results seem to suggest that something other than LTV is influencing the results, which I believe is part of the argument against strict downpayment rules for QRM.

In other words, the chart seems to argue against your conclusion that can’t have a safe mortgage with a low downpayment. In fact, it appears that qualifying criteria beyond downpayment can influence performance, suggesting that, given proper underwriting criteria, you CAN have safe mortgages with low downpayments.

By: TFF Tue, 12 Jul 2011 22:19:58 +0000 “That’s very high, and that single datapoint alone would suffice to show that the 20% downpayment level isn’t a guarantee of safety when it comes to mortgages.”

Many of those loans were underwater within a few years, as the market plunged 30% or more… A 20% downpayment is much safer when there aren’t also millions of RISKIER loans being written at the same time. The riskiest loans soured first. That brought down the market (and economy), causing a cascade across the whole spectrum.

Easiest place to break a cascade is at the start. Eliminate unqualified loans, eliminate low-LTV loans, and the rest would be in much better shape.