Arnold Kling  

If I Were Mortgage Czar

Recession-Proof... Two Essays on Today's Youth...

If I were the czar of the mortgage market, I would attempt to bring back the 30-year amortizing fixed-rate mortgage with a 20 percent down payment. As recently as fifteen years ago, this was the workhorse of the housing market.

I think that a housing market that is based on mortgages with low down payments is inherently unstable. The buyer's equity comes almost entirely from house price appreciation. That means that in a rising market, everybody can buy a home. In a flat or falling market, nobody can buy a home. [note: Bob Shiller wants mortgages that are indexed to home prices. Such mortgages might adapt to this problem, but I would rather avoid the problem in the first place.]

The 30-year mortgage with 20 percent down produced a very stable mortgage market for many years. Until the 1970's, savings and loans provided mortgages in a no-brainer fashion. The joke was 3-6-3: pay three percent interest on savings, charge six percent interest on mortgages, and be on the golf course by three PM. Only when inflation got out of hand did this model break down.

The demise of the savings and loans produced the rise of Fannie Mae and Freddie Mac. One can argue that Fannie and Freddie were more efficient and better hedged against interest-rate volatility than the savings and loans. If Fannie and Freddie had stuck firmly to the 20 percent downpayment mortgage, we would not need a mortgage czar today. But they didn't stick to it, and we are where we are.

I think we ought to aim to go back to the model that worked well in the past, where lots of companies made mortgage loans and kept them in portfolios. That model worked when the 20 percent down payment was the rule. All it would take to get back to that model would be to freeze Fannie and Freddie's mortgage business and to lower capital requirements at banks for purchasing mortgages with 20 percent down payments.

Instead, the plan is to grow Fannie and Freddie's mortgage business for the next 18 months, and then to shrink it. Maybe. Depending on what the next Administration decides to do.

My approach would not do anything to support house prices. In that regard, Shiller and I are in agreement: there is no public policy purpose served in trying to levitate home prices.

My approach would make it more difficult for many families to achieve home ownership, because many families find it hard to save the money for a 20 percent down payment. I am willing to let those families deal with being renters. As fantasy mortgage czar, I value stability of housing markets and financial institutions more than raising the home ownership rate.

I don't expect to be named mortgage czar any time soon.

Comments and Sharing

COMMENTS (20 to date)
dWj writes:

People, a couple years ago, thought buying a home was better than renting because 1) they expected home-price appreciation fast enough that their equity would be increasing by more than they were paying interest on their mortgage, and 2) they tended to compare mortgage payments to rent payments, as though there were no other current expenses associated with owning a home. (I have a friend who has very little financial knowledge but is fairly intelligent, and he believed when buying his condo that it should be *free* -- the appreciation should cancel out any money he spent on mortgage payments, maintenance, property taxes, etc. I suggested to him this was rather optimistic.)

Over long periods of time and broad swaths of geography, there should be no persistent market inefficiencies, and, modulo tax distortions and certain agency costs associated with renting, paying a mortgage and upkeep should cost you, in current terms, about as much as rent plus the increase in home equity that exceeds a normal return on the equity you already had. Here, then, is where I'm going with this: if you can't afford to rent your current place while saving up for a 20% down payment, you can't afford, under normal market conditions, to buy your current place. If you want to trade down, do so in the rental market first, and start saving up; over five to seven years, you can save up your down payment. Buy your house then.

Chuck writes:
Gary Rogers writes:

You may not expect to be named mortgage czar, but you certainly would get my vote.

What you say makes so much sense. The only problem is that if we go back to the 20 percent down model now it will depress home values for the next 5 or 10 years while potential homebuyers accumulate enough cash for a downpayment. Meanwhile, even existing homeowners lose their equity due to the drop in prices and are locked into an upside down mortgage. The moral of the story is that once you follow bad advice, it is hard to go back.

Mercutio.Mont writes:

Doubt that the government can do much to maintain inflated housing prices in a country with as much empty space as the US. There is so much room to build, no reason for houses to stay expensive.

China Expat writes:

You have to put down 30% on new apartment in China, and 50% down on second-hand. All made to discourage real estate speculation. And prices are still going up.

Prakash writes:

Why not islamic co-ownership mortgages?
The mortgage company and the person taking the loan buy the property together. Every year they decide on the value of the property and the rent to be paid. If the property is 10000 USD and the bank foots 9000 USD and the buyer 1000, then the buyer has a 10% EQUITY. The buyer pays 90% of the decided rent, since he owns only 10%.

Everytime the buyer pays more than the decided rent, that goes into increasing his equity. Everytime she falls short, that comes off the equity. The company and the buyer swim and sink together as the property value increases and decreases.

It is a fair system that should be considered in the western world, as it seamlessly transitions between renting and owning.

Gary Rogers writes:

To China Expat:

It is not the level of the down payment that determines whether prices will rise or fall, it is the transition from one level to another. We would be doing fine if everyone were expected to put 20% down and this remained constant. But if this is suddenly dropped to 10%, prices rise as more buyers enter the market. Conversely, if the downpayment has stabilized at 10% and is suddenly dropped to 20% prices will drop for a time while buyers are forced out of the market for lack of a down payment. Doctor Kling is absolutely right that 20% is a much better level, it will just be painful deleveraging the market and may topple some additional dominoes.

Aaron writes:

While I agree with a lot of what you said (lower LTVs, no reason for the gov't to prop up prices, etc), I don't follow your hostility toward securitization. Yes it's been abused over the past decade, but I think we still want to boost liquidity and encourage the spreading of risk. If not, and the banks hold the loans, we freeze up capital and make any necessary future reallocations much more difficult.

Barkley Rosser writes:


Clearly things would be more stable if a lot of the sorts of mortgages that were being issued in recent years were not issued in the future. However, I see no reason to eliminate 15 year mortgages, which involve higher monthly payments and a more rapid payoff, essentially involving increased savings. Also, it might be wiser to be a bit looser on that downpayment, maybe 10%.

It is a bit cavalier to note the rise of the Agencies after the massive wipeout of the S&Ls at the end of the 1980s. The Agencies had been around a long time before that.

The problem for them has been the pressure to loosen their standards, which with respect to backing subprime mortgages they only got around to doing in late 2005. That was because pretty much all their fully private competitors were doing so, making it very hard for them to hold back the practice, although it would have been better if they had been able to do so. But they no longer really had control of things by then, even though after things blew up last year, all those private secondary mortgage players disappeared from the market, leaving the Agencies holding the bag.

John Thacker writes:

Mr. Rosser:

"It is a bit cavalier to note the rise of the Agencies after the massive wipeout of the S&Ls at the end of the 1980s. The Agencies had been around a long time before that."

Yes, but their market share certainly rose after the S&Ls fell, concentrating the market. That's worth noting.

"The problem for them has been the pressure to loosen their standards, which with respect to backing subprime mortgages they only got around to doing in late 2005."

You're confusing the issue here, Mr. Rosser. Even if Fannie and Freddie didn't loosen the standards on mortgage loans originated by them until late 2005, that's only a small part of their portfolio. They were repurchasing subprime mortgages in a big way from Countrywide, WaMu, and the rest back in 2002 and 2003 as well. By repurchasing those loans and providing guarantees (and successfully lobbying Congress against restrictions proposed by the Administration), they certainly boosted the subprime market.

Dan Weber writes:

I didn't see Arnold's call for a return to the 30-year mortgage to mean eliminating the 15, but rather everything more risky than the 30.

The 15-year (with 20% down) is less risky than the 30 year.

Lord writes:

That would be overly drastic. This could have been prevented by doing only one thing: insuring people could pay back what they borrowed. That tethers prices to incomes and prevents appreciation from getting out of hand. Good credit or bad, high down or low down, fixed or adjustable, rich or poor doesn't really matter; all that matters is they can afford it and if they are not willing to pay for it, the next owner can. That means requiring proof of income, and qualifying people on fully adjusted market rates not teasers.

Ross Wilder writes:

As one whose knowledge of such matters would have trouble filling the head of a pin, I pose the following for hypothetical purposes only......

Why should housing require a 20% down payment when very few other 'credit' purchases do? E.g., there's a decent chance I would need to purchase a riding mower to help maintain the yard of a house on which I just put 20% down. A mid-range John Deere rider is likely to cost me somewhere in the $15-1600 range. But like as not, the mower purchase would require no down payment whatsoever.

Why should one purchase require the DP, and the other not?

Of course, playing devil's advocate to my own question, perhaps the best answer is that *all* credit purchases should require a D.P.(??)

Or alternatively, perhaps everything *except* housing should require X% down.(?) Though the current 'crisis' revolves primarily around housing, would the situation be nearly so dire if the avg American weren't so highly leveraged across the board???

Hopefully, this isn't too simplistic a question for the more knowledgeable in the crowd.

RL writes:

Ross W:

IANAE, but at first approximation, I'm guessing you don't need to put 20% down on a riding mower because,

a) it's a lot easier to repossess a riding mower than a house
b) it's a lot less socially destructive to repossess lots of riding mowers than lots of houses (unmown lawns vs families dispossessed).
c) society's investment in housing stock is a lot higher than society's investment in riding mowers, so screwing up there is more socially destructive. There is no riding mower equivalent to Fannie and Freddie...

Superheater writes:

The government has accepted the default risk on different loans for years, at least in the education loan market.

Any state government operated higher education assistance agency that attains "exceptional performer" status automatically transfers all of the default risk in the education loans in its portfolio to the federal Department of Education. If the borrower defaults, we pick up the tab.

The only way to enforce prudent lending practices, is to ensure that the "originator" is in the instrument for the long haul and makes decisions based upon sound economic principles.

Unfortuntely, those things won't happen. The genie is out of the bottle with securitization. Banks will never give up this tool of liquidity and earnings management. (Governed by FASB 91, banks are required to defer recognition of upfront lending fees, until the mortgage is sold, so if earnings are a little flat, sell some loans and recognize some of those deferred loan fees.)

Secondly, the minimal lending scrutiny wasn't just the result of some collective banking insanity. No income verification, no collateral mortgages were the direct result of "anti-redlining" legislation, such as the CRA. Obviously, you can't be accused of discrimination if you just don't collect information or impose requirements (such as collateral) that might positively correlate with race or some other protected attribute.

As for the idea that there's something magical about 30 year mortgages, I don't think so. A 15 year mortgage builds equity faster, involves less risk of default due to morbidity, disability or the "vissitudes of fortune".

Derek Scruggs writes:

As a non-economist, it seems to me it would also help for the government to better calculate housing costs in the inflation rate, which I believe are currently excluded or understated (someone please correct me if I'm wrong). Here's an analysis from 2006 I just found via Google.

I first encountered this idea couple months ago in an op-ed piece, though I can't remember where. Now that I think about it, I don't know if a higher reported inflation rate would suppress prices or spur more buying as a hedge.

John Thacker writes:

Derek Scruggs:

The following link is to the best defense of how the CPI measures housing costs (and everything else), by two BLS economists, including a Division Chief, Robert McClelland. The housing section is the part entitled "Rental equivalence" on page 10 (8th page of the PDF).


In short, there are very good reasons why rental equivalence is used. That means that home prices are valued for terms of inflation in what a comparable home would cost to rent, not the cost to buy. (Buying includes a lot of investment in addition to consumption, so it conflates things.) Note that your Internet article is from 2006, when home prices were high and rising and including asset values instead of imputed rent would have raised the CPI. By comparison, including asset values now in a time of falling home prices would make the CPI much lower, not higher.

Dan Weber writes:

This could have been prevented by doing only one thing: insuring people could pay back what they borrowed

Where is PMI in all this mess?

Are the PMI insurers going bankrupt? Did the brokers just sell the loans without getting PMI?

Brock writes:

Bah. One size does not fit all. There can be legitimate reasons for not putting 20% down. There can be legitimate risk takers out there willing to lend to a less-than-20% borrower.

The current problem's source is twofold: (1) non-pricing of risk, and (2) not anticipating unlikely events. The first was created by lax lending standards - income was not verified, homes were not inspected, etc. The second was caused by the belief that market events are normally distributed rather than power law distributed.

Arnold's suggestion will work because it simply outlaws risky lending and creates enough padding to survive most disruptions. But that doesn't make it optimally efficient. Further, modern Wall St types prefer making their year-end bonus to being on the gold course at 3 PM so a boring, liquid, well understood market will not get them out of their chairs.

Lord writes:

PMI largely became obsolete, displaced by piggyback loans. The 80% loan didn't need it and the 20% loan had a higher rate to compensate. That would have worked just as well if prices were not allowed to detach from incomes. We certainly never had a massive swing in incomes.

Comments for this entry have been closed
Return to top