Closer Look At The ARMs Reset Problem

There's a detailed set of slides breaking down the woes in the housing market out from Real Estate Consulting. Although it's an interesting report, a pair of back to back slides from that presentation appears to be off the mark.

Here are the slides in question. A discussion of those slides follows.



click on chart for sharper image



click on chart for sharper image

The first chart states "ARM loans made in 2004-2006 will create huge reset problems in 2007-2009". The second chart specifically points out loans made in 2005-2006.

Let's start with a discussion of Adjustable Rate Mortgages [ARMs]. ARMs are based on an index rate, typically 1-year treasuries or 1-year LIBOR, plus a margin amount (e.g. the treasury rate + 2.75%) .

The interest rate is fixed for an initial period (3 years for a 3-1 ARM, 5 years for a 5-1 ARM, etc.) but then floats with the index, adjusted periodically (typically yearly).

The margin amount varies lender to lender and it does not change when a loan resets.

With that background, here are charts and tables of treasury rates and LIBOR for the last 10 years. We will use the rate tables to see what's likely to happen when mortgages reset.

One Year Treasury Rates

click on chart for sharper image

One Year Treasuries Table



click on chart for sharper image

One Year LIBOR Table



click on chart for sharper image

The previous three charts courtesy of Money Cafe.

An important word about the above tables:

Initial rates on any given day may be based on factors other than the index rate at that time. In cases where lenders were aggressive, initial rates may have been lower than implied above, creating more of a reset problem down the road. In cases where lenders were more cautious, initial rates may be slightly higher than implied, creating a more beneficial situation when rates reset.

Specific circumstances vary day by day and lender to lender. Thus, the above tables are best used as a guideline as to what took place, as opposed to an absolute mathematical reference point.

Current Quotes as of March 29

• 1 Year Treasuries - 1.56
• 1 Year LIBOR - 2.52

3-1 ARMs Analysis
Based on the above table, 3-1 treasury based ARMs initiated in 2005-2006 would likely reset lower. I stress likely because initial rates on any given day may be based on the above caveat on day to day conditions, lender specific conditions, etc.

3-1 LIBOR based ARMs initiated in 2005-2006 would also likely reset lower and again with the same caveat repeated about day to day conditions, lender specific conditions, etc.

In general, loans originated in 2005-2006 simply do not appear to be a problem. In fact, the reset of 3-1 ARMs from those years may provide economic stimulus. This statement may not apply in every case but it should be true for many, if not most cases.

3-1 ARMs prior to 2005 have already reset. So those problems, whatever they were, have already been faced.

5-1 ARMs Analysis

5-1 ARMs analysis is more difficult. However, we can say that the above charts clearly show that loans originated in 2003-2004 are more likely to be problematic than loans originated in 2005-2006.

Here is an analysis from my Certified Mortgage Planning Specialist friend:

"Most of the 5/1 ARM’s I reviewed that originated in 2003 had start rates between 4% and 5.125%. This makes it hard to lump them all together to state they will in aggregate reset higher or lower. Some of this will depend on points paid to reduce the original interest rate. Many of these loans will reset marginally higher than they were before. However, the situation is far better now than it was even a few short months ago where it appeared virtually every loan in this group would reset much higher."
While likely to be net negative to the borrower, 5-1 ARM adjustments from 2003-2004 are not likely to be the end of the world. Some may even benefit. Looking ahead, adjustments on 5-1 arms for 2005-2006 are likely to be consumer friendly based on the above tables.

However, 5-1 ARMs from 2005-2006 will not reset until 2010-2011. Those loans are simply not today's problem.

Principal Payments Needs To Be Factored In

There is still one more issue to address, and that is higher payments when the interest only period ends. For example, a 5-year ARM loan typically goes from interest only payments to interest + principal amortized over 25 years on the first rate reset. Likewise a 3-year ARM loan typically goes from interest only payments to interest + principal amortized over 27 years on the first rate reset. Some ARMs have a 10 year interest only period which postpones this particular problem.

Many of those in 3-Year ARMs with principal and interest payments will see their total mortgage payment drop. This is especially likely for loans originated in July of 2005 or later. However, those paying interest only for three years may see their total payments rise.

For others, especially those in 5-year ARMs, there could be payment shock even if the interest rate drops. Once again, it will be the 2003-2004 loans that will prove to be more problematic.

Where Is The Reset Problem?



click on chart for sharper image

Poof!

The Fed vaporized much of the ARM reset problem in 2008-2009 by slashing interest rates. This is especially true for those in 3-1 ARMs. It is highly likely the Fed's panic rate cuts in early 2008 were made with this in mind.

Looking ahead, and assuming rates stay low, those in 3-1 ARMs originated in 2006 are likely to see significant resets lower. This would be economically stimulative as long as other conditions are stable.

Unfortunately, other conditions will not be stable. I believe unemployment is going to soar, commercial real estate is going to plunge, and the stock market is likely to keep heading south. Given rampant overcapacity everywhere, I see no economic force to create jobs. All of this is going to further pressure housing prices and the economy in general.

Look Ahead to 2010-2011

Looking ahead to 2010-2011 I see a different set of problems.

Those problems are Alt-A and Pay Option ARMS. And that is where the liar loans (no-doc loans) are hidden. Liar loans are likely to blow up long before we get to 2011. I discussed a particular Alt-A pool in WaMu Alt-A Pool Revisited and WaMu Alt-A Pool Deteriorates Further.

The pool discussed above originated in May 2007, and was 92.6% rated AAA. The most recent update shows the pool is already 25.3% 60 day delinquent or worse, 13.35% in foreclosure, and 4.44% REO (Real Estate Owned). The problem is easy to spot: only 11.27% of the pool had full doc. The rest of the pool was liar loans. The pool may not be representative a representative sample of Alt-A pools. However, it does illustrate the type of problems one would expect to see with liar loans. And those problems are both big and growing.

Pay Option ARMs [POAs] pose additional problems. The first problem is that over 80% of POA mortgagees only make the minimum payment. Given that minimum payments typically do not cover interest owed, the loan balance increases every month. This is called negative amortization, and it has been going on for years.

Negative amortization is compounded by falling home prices. At some point, typically 110-125% of the mortgage, an enormous gotcha kicks in. That gotcha requires a fully indexed fully amortized principal and interest payment, amortized over the remaining years. People who could only afford the minimum payment will be forced to pay principal, plus interest, on top of a loan balance that has been growing monthly. Good luck on lenders getting all their money back on those loans.

The second problem in regards to POAs is that a huge portion of these loans originated if the least affordable, biggest bubble areas, like Florida, California, Las Vegas, etc. From a lender's perspective that hugely increases the likelihood of default as well as the size of the problem should default occur.

Mortgage Reset Scare

There are lots of mortgage problems for sure, but right now loans resetting from 2005-2006 do not appear to be among those problems. Nor are 3-1 ARMs in general a problem. 5-1 ARMs specifically from 2003-2004 are more problematic as noted above, but even then, lower interest rates on which those products are typically aligned have alleviated some of the concern.

However, people are worried, especially after hearing about the mortgage rate reset time bomb for a number of years.

Unfortunately, there are companies attempting to take advantage of the mortgage reset scare by offering "special rates" to refinance now. Those "special rates" are in cases much higher than where they would automatically readjust to.

I discussed this sad situation in Dear Citigroup Customer ....

If you have an ARM about to reset, or know someone who is in that situation, please pass along the above link. There is no excuse for major banks to be attempting to take advantage of distressed borrowers, whether it is legal or not.

There was a very nice discussion of the Citigroup mortgage rate reset issue from an operational standpoint on the Practical Risk Management blog.

Inquiring minds may wish to read Operational Risk – Improper Disclosure By Citigroup Mortgage. The article discusses a potentially serious breach of fiduciary responsibility by Citigroup, possible RESPA violations, potential violations of Reg. Z, and likely violations of internal procedures.

Anyone who feels aggrieved by the actions of Citigroup (or any other lender) may wish to contact http://www.consumergripes.net/

Comments

[Maniac inMotion]

I think the article is right on with the 2 major problems: ARM Resets and "Limited or No Doc" liar loans. However, I think liar loans will go to foreclosure now and ARM resets will go to foreclosure slowly through 2011. Here's why.

Buyers of liar loans were predominantly speculative investors or self-employed business types who were after one thing: ROI. With a keen understanding they are upside down, and the likelihood they've likely already done well with previous trades, they have no incentive to hang on to a property where they might be 30, 40, 50,000 in the hole in an immediate time horizon. These types will likely make a conscious decision to stop making moretgage payments on a house they can't flip to cut their losses.

Meanwhile, some folks with ARM'a resetting were unfortunately duped to buy their home with an ARM instead of a fixed rate mortgage, and will likely slowly fall to foreclosure. They will try their best to keep the one home they dreamed of buying, but if house prices don't recover in time, will fall to foreclosure.

Sooner or later, one thing will have to give. The Fed will either see no choice but to raise rates, or we'll see the greatest hyperinflation in American history by keeping rates low like Japan did in the 90's. I imagine Bernanke is VERY AWARE of what repercussions would be experienced by keeping rates too low for too long AGAIN. I think we may likely see Fed hikes immediately following the election.

With Fed rate hikes, we'll see the 1 year Treasury rise, once again hiking up mortgage rates on folks with ARM's. Homeowners will likely have no choice but to foreclose if they are still upside down on the home, which they likely will be upside down. The consensus is it would take 10 years to see home prices recover. This will further deflate values from the 2006 highs.

To sum up, I think the article's right on the 2 major housing problems. But I think it's backwards. Liar loans are the immediate problem today, and folks with ARM's resetting will be the lingering problem tomorrow.

[cfish]

I'm not big into litigation and interventions but citigroup needs to pay for trying to scam people.

[cfish]

Good observation! I would like to add my 2c about possible prolonged effects that will drag on forever.

It takes 90 days to go through a typical foreclosure process for a home to be REO. After that, typically, banks hire someone to "clean it up" and list it with a Realtor - which then will sit on a market of over 10 months of supply to move. Most banks probably have too many foreclosures to go through all these stages. What will they do? They have many options, and I believe whatever they might do will end up making the house listed cheaper on the market.

Then the snowball starts rolling for real. More mortgages will be underwater, making more people walk away.

[PublicLiterature.org]

This is a very thorough article. I enjoyed it, but I believe that the 2011 problem may be overstated. I believe home equity should be built up by this point (home values may be near all time highs 3 years from now). This would only require a market stabilizing and perhaps increases in value 2-4% from 2009-2011.

Even if home prices remain at their current levels, the Fed understands the problem so well, that they will jump in again and be more accurate with their actions.

[braggbear]

Unfortunately, I don't think this article proves anything. Remember: The difference in monthly payments from before the reset to after the reset is not just a function of initial rates and current rates; it is also a function of the fixed margin amount. So even if the current rate is below the initial rate, if the margin is high enough, the payment will go up.

Until I see some data on the average and distribution of margin amount for loans vintage 2003 to 2005, this argument will remain unsettled.

[fluffymcdeath]

Troubling enough of itself, but there is a whole house built on this foundation of cards.