Mortgage Down Payment Size vs. Delinquency Rate

As part of new reforms, the government is debating what downpayment size should be required for a “qualified” residential mortgage. If a mortgage doesn’t meet the new standards, the lending bank would have to retain 5% ownership even if selling the rest to investors. The initial proposal is for 20%, but of course the mortgage industry wants the required down payment to be as small as possible. They want to keep the good ole’ days from being able to offload the risk entirely onto others.

Honestly, if banks can’t handle keeping even 5% ownership of the loans they originate, why would I trust their underwriting at all? Their track record for determining creditworthiness hasn’t exactly been stellar. Of course, their public argument is that a low downpayment keep homes “affordable” for everyone. From the Washington Post:

“Why, in a law intended to fix the mistakes that caused the credit crisis, would you mandate a certain down payment when low down payments were not the problem?” said Kathleen Day, spokeswoman for the Center for Responsible Lending.

Actually, they are a problem. Felix Salmon points out how the mortgage industry is trying to influence people with misleading statistics, saying that “boosting down payments in 5 percent increments has only a negligible impact on default rates.” After some wrangling, Salmon got this clearer chart showing delinquency rates as a function of downpayment size for the period 2002-2008:

When the mortgage industry starts complaining about the 14 million people who would be denied the chance to buy a qualified mortgage if they don’t have a 5% downpayment, it’s worth remembering that qualified mortgages for people who don’t have a 5% downpayment have a delinquency rate of 16% over the course of the whole housing cycle. (You can be sure the numbers were much higher still in 2006 and 2007, which is why Guarino didn’t give them to me.)

And you can see too why the 20% downpayment limit was put in place: it’s the point at which delinquencies fall to less than 5%. If you take one group of loans with a 20-25% downpayment, and a second group of loans with a 15-20% downpayment, then the second group, on these numbers will have a delinquency rate 56% higher than the first.

The fact is, downpayment size does matter. Imagine what the chart above would look like with data from 2006 to 2009. It should serve as a reminder that giving anyone with a pulse a mortgage loan because it’s the American Dream was a bad idea.

A 20% downpayment was the standard when banks actually kept 100% of mortgage on their books. I’m not saying every single person should need 20% today; A bank should be able to create a mortgage that requires less, but in exchange it should have to have some skin in the game. If a lender won’t even keep a mere 5% on their books, doesn’t that just show the mortgage is too risky in the first place?

More reading: NY Times, ChicagoMag

Comments

  1. I actually think that everyone should have to put down 20%. If they don’t have 20% in cash they should put down 10% and borrow the balance as a second lien against the house with personal recourse. If they can’t get approved for this second lien they should keep saving. Maybe that is too harsh for really expensive housing markets, but then again if those were the rules maybe some of the markets wouldn’t be so expensive.

    However, let’s be clear about the 5% requirement for banks. I don’t think the banks are required to strictly hold a 5% slice of the mortgage. I think the requirement is that they hold 5% of the origination in the equity tranche of the securities they are selling. There is a big difference. Holding a 5% vertical slice of one or all mortgages entails a relatively low level of capital risk, because as you pointed out even the riskiest loans have a sub-20% expected default rate, and even if severities were 50% they would still only lose around 2.5% of their investment in the very worst housing market imagineable. However, if they own a 5% equity stake in the RMBS they are in the first loss position and it takes a much lower default rate for them to lose serious equity, which is a big no-no for the banks.

  2. Baughman says:

    I like Andy’s comments. What I find comical is that 95% of the population has no ability to refrain from spending to accrue this elusive “savings.” Houses are the only accidental savings accounts for many of the people that I know, provided that they haven’t broken the piggy bank with HELOCs.

  3. Borrowers should be able to seek loans for any amount and LTV ratio. Lenders should be able to set their own LTV requirements for loans. The problem isn’t that borrowers are putting too small of an amount down, the problem is that the government is insuring loans with 3.5% down payments. If a lender makes a risky loan and the borrower defaults, the lender (not the US taxpayer) needs to bear the consequences for making a poor decision.

  4. I have to disagree here, the numbers given are too general too draw a meaningful conclusion from. One reason is that they aren’t correcting for a host of other loan attributes, higher LTV loans have a much higher incidence of fraud for examle, that could easily explain the default rate differential instead of downpayment.

    It has been my experience in the distressed mortgage trade that the major factor predictive of default rates is current loan to value. Borrowers default at relatively loan rates (even today) when their LTVs are low, when LTVs rise default rates spike and they do it fast. This jump in defaults doesn’t happen when you go underwater either, it happens at around 110-115 LTV. This makes sense, defaulting on an underwater house does cost you in time, moving expense, and a credit score hit. The gain from defaulting has to be greater than your cost so you will be willing to go slightly underwater. In this way low downpayments correlate with higher default rates, but that is only true because of the large drop in housing prices. In markets with flat home prices the default rates of small downpayment loans is not markedly different than those with large downpayments. Practically that means that low downpayment loans can be made profitably, that is they can charge reasonable interest rate spreads that will exceed expected losses on the loans. To effectively eliminate low down payment loans would be harmful to the nation.

    Look at this scenario for example, these are real life current market values. Currently in central and western Florida homes are selling at a 5x cap rate, that is they sell for 5 times what they can be rented for annually. That means a $1650 rental sells for 100k, financing that home with a 2.5% downpayment at 5.5% would cost an owner $550 a month. You can’t possibly argue that there isn’t a large market of people for whom the 2.5% downpayment low mortgage payment ownership option is far wiser financially than to rent for $1650. Forcing them to accumulate $20k while paying an extra $1100 per month to an investor is unnecessary.

    I’m not saying that capital requirements on loans don’t need to be revised, they do. But using a 20% downpayment as a threshold, like in the example above, leads to obviously reasonable loans being disallowed and borrowers being harmed.

    One suggestion I’ve made is that low downpayment loans be required to have accelerated principal amortizations until their LTVs reach 80%. For example a 5.5% rate 97.5 LTV 30 year loan will take almost 11 years to reach 80 LTV through normal principal amortization. Perhaps you say that any loan above 80 LTV must contain larger scheduled principal payments that will bring the LTV under 80 within 5 years of origination. Using the 100k house with a 97.5 LTV loan mentioned above as an example you can see this still would be attractive to homeowners. A bank could charge 8.5% (an incredibly juicy spread) and require $400 in extra principal payments each month. That would make the monthly cost of the loan $1150, a savings of $500 off rental rates, and bring the loan below 80k principal in only 3.5 years. After that the $400 principal payment is no longer needed, they can refi to a more reasonable rate, continue their accelerated rate, or just pay the minimum $750 per month. Rents are high, home prices and interest rates are low, there is a shocking amount of room to concoct reasonable low down payment mortgages.

  5. Totally agree with the last part you described Bill. You’ve basically described a second mortgage for the down payment to get the first lien LTV down to 80% with a 5- or 10-year amortization and higher rate. I would add that that second lien should be recourse to the borrower (not just the house) to prevent the homeowners from having a free option to walk away that comes along with their lower down payment mortgage. Those free options create problems when LTVs rise as you described in your falling price scenario.

    The reason we can’t go the full free-market route of allowing lenders to make any loan and try to price them correctly is because we have (for very good reasons) deposit insurance and a lender of last resort to promote stability in our banking system. The trade-off of eliminating the likelihood of having a run on the banks is that the banks no longer are fully responsible for the loans they make. If enough of their loans go bad the losses reach the depositors. The depositors don’t have an interest in underwriting their banks because their deposits are insured. If we are going to have a fractional reserve system we need a lender of last resort to keep it from collapsing at moments of lost confidence, and if we have deposit insurance we need rules to prevent moral hazard among the mortgage lenders.

  6. @Bill – I like your suggestion. I realize you’re suggesting more of a hard and fast rule, but I just wanted to point out that the FHA rules already push borrowers in the direction of accelerating principal payments along those lines the first five years.

    FHA Borrower pays mortgage insurance for at least 5 years until they reach 78% LTV. Hence the way for an FHA borrower to get off the mortgage insurance as soon as possible is whatever additional principal payment walks you down the amortization table to 78% LTV at exactly the 5 year mark.

  7. It’s similar but I described it as a single loan because it’s so exorbitantly expensive for the 2nd that it would run afoul of usury laws. A 5% and a 20% loan might not fly while a single 8.5% loan would. Also structured as a single loan you avoid the problem of 2nd lien owners holding the 1st lien holder ransom when they are completely uncollateralized.

    The problem with mortgages is that all the rules are decided at the state level while bank capital rules are done at the national level. In many states, even the infamous California, 2nd liens are full recourse. There it’s only 1st lien purchase loans that are not. I suppose you could have the capital rules differ depending on the legal treatment of each particular loan.

  8. I think a 20% down payment makes sense. This will not affect affordability as Kathleen Day says as home prices will grow slower if borrowers need 20% to buy. If 20% isn’t required, I like Bill’s idea of a faster amortization to get them to 20%.

    I don’t agree with Andy about not being able to go to the full free-market route. Deposit insurance should, in theory, be covered by insurance premiums. The reality is the government doesn’t run FDIC that well to cover a big default, but it should. As far as the Fed, it is anti-free market in nature as it interferes with the markets. The Fed is the primary reason we had the housing bubble so I think a case for a true free market is strong. Part of that market working well is borrowers having more skin in the game.

  9. Agreed Dan, releasing the mortgage insurance at 78 LTV does provide incentive for a homeowner to build equity. I’d like to see the cost of higher LTVs even higher and thus the benefit to early equity building be that much greater.

    Another option I like would be to index your rate monthly to your LTV. Five years is a long horizon and a lot of people have trouble paying now for a benefit five years away. If you indexed their rate monthly they could tangibly see a difference with every single principal payment. Maybe they owe 5.5 + (LTV-80)/20 * 3 on the entire balance with a minimum of 5.5%. That way every month they can watch their interest rate fall as they plow in extra principal.

  10. its fair enough for homeowners though I hope it is a little higher than the usual.

  11. @ Bill– Wow, where do I start. I see several problems with your insight and proposals starting with your first post:

    You are picking one area of the country where you claim that housing payments are one third of what rental costs would be (the other two thirds being the hole the renters go into each month. This scenario if it is as you say has just been created because of all of the defaults of people who obviously could not afford their homes and/or were willing to walk away because it suited them. So now home values have been greatly diminished and all the foreclosures have left a shortage of available housing causing rents to rise to this level.

    First, if someone would have to rent at three times the cost of a house payment–because they can’t purchase due to no savings–obviously (to me I guess) they should be renting some place much cheaper while they save $ (not renting the house of their dreams).

    You also do not mention the outlandish expenses investors and homeowners must pay for hurricane insurance, homeowner’s insurance, high property taxes, major fixes that come up, etc, etc, which the investor or homeowner must pay. These are also problems that have been the downfall of low DP homeowners who “forgot” to plan for the bad times and the extra expenses of homeownership.

    You say, “To effectively eliminate low down payment loans would be harmful to the nation.” I would argue just the opposite. As a country, we would be much better off requiring larger down payments. When people prove that they can save a significant down payment they are showing their willingness and good sense to be able to afford not only the large down payment and the ongoing house payments, but any significant “surprise” expenses that will inevitably come up when they may or may not expect it. How can that make more sense?

    And, because of more stringent ownership and DP guidelines it should prevent the artificially high home prices that we saw heading up to the 2006/2007 housing price collapse. Wouldn’t that be more of a win-win for everyone. After all, what good does it do anyone to have someone qualify for a low DP loan only to lose it all and have their credit trashed when they find they bit off more than they could chew?

    And the initially higher payments for the first few years along with the higher interest rate to compensate lenders will likely just fuel more foreclosures earlier in the homeowner process. Again, how does that help anyone?

    Jack

  12. I think I did not explain my point carefully enough, Jack, because several of your points don’t apply.

    The situation in Florida is as I described. “So now home values have been greatly diminished and all the foreclosures have left a shortage of available housing causing rents to rise to this level.” This is not the case. There is not a shortage of available housing at all, in fact the home vacancy rate is the highest in the nation, 17.5% of homes in Florida were vacant at the end of 2010. How can there be a huge housing surplus and such a massive price/rent ratio discrepancy? It’s simple, there is a large shortage of borrowers who cannot meet new more stringent underwriting standards, one of which is a requirement for larger downpayments. I’m not in anyway suggesting the solution is to give anyone a loan who applies, I’m only saying that some reasonable borrowers are now excluded from the mortgage market.

    It is also the situation in suburban California cities, Arizona, Nevada, the rust belt states and other areas. In suburban Massachusetts a relatively stable real estate market has rents currently pricing at about two times a mortgage payment. I obviously picked a very distressed market as an example but the exact numbers aren’t particularly important, I was only trying to show that there is a massive amount of room between rental and ownership costs. Even with insurance, property tax, etc there is a lot of room for ownership to be a better financial decision than renting.

    I think we actually agree a lot more than we disagree on the subject of housing. The one point I didn’t make clear earlier was the importance of underwriting which I think you also have missed. You say that the downfall of low DP borrowers is that they forget to plan for bad times and they can’t afford the extra expenses of homeownership while large DP borrowers will be able to afford significant “surprise” expenses. These factors may be correlated but they have no causal connection. The person with 20k in the bank can buy a 100k home with an 80 LTV mortgage and be less prepared for surprises than the guy with 10k who puts down 5%. The guy who takes in 1k a month to pay his $500 mortgage hasn’t planned for extra expense while the guy making 2k for his $600 higher LTV mortgage has. You can separate out these cases with proper underwriting. My point is these examples are not uncommon, there are a tremendous number of high income borrowers who are good credits who have not had the time to save up a 20% downpayment. I’d be perfectly happy if 80% of applicants are denied for low DP mortgage because it means that the 20% of qualified borrowers can now enter the housing market.

    It seems that you aren’t considering the loan underwriting process properly, your last comment makes this clear. “initially higher payments for the first few years along with the higher interest rate to compensate lenders will likely just fuel more foreclosures earlier in the homeowner process”. This is simply not true, higher required payments will have no effect on default rates. Instead they have an effect on who gets a loan at all. If you offer a loan type with higher initial payments you will simply price some people out of the market, those who can still borrow will qualify for smaller loans. I have no problem pricing low quality borrowers out of the market, in fact this has already occurred to a massive extent with the collapse of the non agency mortgage market. I’m simply saying that LTV alone is a bad way to measure loan quality. 80% LTV maximums price out a lot of good borrowers. I have no problem saying that a simple 97.5% maximum will price in too many bad borrowers also.

  13. LargeTalons says:

    Why should we regulate LTV at all? Shouldn’t a lender be able to make a loan using their own best judgement as to the risk/reward? I actually think LTV maximums are part of the problem. What do you think happens when someone who either foolishly bought into or was fooled into buying a risky loan tries to refinance out but finds out their LTV is now too high? They are stuck, and they get foreclosed on. If an investor doesn’t trust a company’s underwriting they don’t have to buy their loans. All this mucking about in the market by the government helped cause the very problem they are trying to fix by more regulation. I’m not saying buying a house, or giving a loan to someone with less than 20% down is a good idea, but I think two consenting parties should be free to enter into that agreement if they want. We need accountability and freedom, not nanny state regulation. I’m not even normally a conservative, in fact, now that we are already in the problem, I would recommend government intervention in the form of subsidies right to home owners. I would propose something were Freddie and Fanny buy any refinance loan at any LTV for any residential property purchased before 2010 at like 3% or something. Leave that open for a year or so and then shut them down to any new investment and phase them out. As a compromise for the banks, make no-recourse laws illegal at the federal level. No-recourse is a ridiculous system anyways. Why do consumers care if they take on too much debt if they can walk away? And why would lenders care not to lend to them if they can just get federal insurance or a bailout? Capitalism without risk is not capitalism. Its reverse socialism.

  14. Why should the government be involved at all? They wouldn’t care as much if they didn’t insure the mortgages!

Speak Your Mind

*