Markets fell on Tuesday, on fears, we are told, about weakness in mortgages and housing; this prompted Brad Setser to ask me whether I’m still as sanguine as I used to be on such matters. I couldn’t reply on Wednesday, since I went to Disneyland instead — I can highly recommend the Indiana Jones ride, if you find yourself there for whatever reason. While I was there, the market went back up again. As ever, my main conclusion from all this is that no one should ever draw any conclusions from one-day market moves.
All the same, I might as well get down here for the record what I think about mortgages and housing, because although I think that the bears have some enormous holes in their arguments, that doesn’t necessarily make me bullish.
I basically start with the observation that the run-up in housing prices in recent years has been seen all over the world: from Spain to South Africa, from Ireland to Australia. Not everywhere has participated: there’s been a lot more price appreciation in Shanghai than there has been in Stuttgart. But there’s clearly some kind of global force — let’s call it “liquidity”, for lack of a better term — which has been driving asset prices in general, and housing prices in particular, upwards. What’s more, by the standards of many of these countries, the run-up in US home prices has been relatively modest.
Let’s now take another look at the US. I don’t deny for a minute that the number of defaults on subprime mortgages, especially subprime mortgages of the 2006 vintage, is eye-poppingly high, and rising. What’s more, given the fact that originators have tightened up their underwriting standards substantially (and correctly), many mortgage holders are going to find it hard to refinance their adjustable-rate mortgages when they reset. This could drive default rates even higher, if the rates on those mortgages go up a lot. Home prices are still substantially above their 2004 levels, however, so the real problem is largely confined to mortgages originated in 2005 and 2006.
Now, it’s easy to paint a picture whereby defaults on recent-vintage mortgages cause a more generalized tightening in credit markets along with a plunge in housing prices as supply goes up and demand goes down. I do accept that, in theory, this could happen. But before I believe the prognostications of those who say it will happen, I want to see some indication that their model of supply and demand and housing prices actually reflects reality. Specifically, I want to see whether it explains the rise in housing prices over the past few years.
The big mistake that I think a lot of the housing bears are making is that they’re confusing correlation with causation. Yes, subprime credit has increased substantially in recent years. And yes, housing prices have gone up. But that doesn’t mean that the former caused the latter. Indeed, there are a good reasons to believe that the increase in subprime credit did not cause the run-up in housing prices. For one thing, housing prices started rising before the increase in subprime credit. For another, much of the increase in subprime credit was concentrated in depressed areas such as Michigan and Ohio, which accounted for 15% of all US foreclosures in January. And it’s precisely those areas which did not see a run-up in housing prices. Meanwhile, the market in New York City co-ops, for instance, has been booming (and shows few if any signs of slowing down) despite the fact that because those units are “non-conforming”, it has always been incredibly difficult to get a subprime mortgage on them. Finally, the rise in subprime credit in the US was not matched by a rise in subprime credit in places like South Africa and China, which saw much greater price increases — so whatever forces drove housing prices up globally are probably sufficient in themselves to explain the increase in housing prices in the US.
Now, one of those forces is doubtless an increase in the availability of credit generally, if not subprime credit in particular. If housing prices rose on a tide of global liquidity, then they can fall if and when that tide starts going out. For the time being, however, there are no signs that is happening. Yes, there are problems in one localized sector of the US mortgage market. But credit markets globally are still as tight as they’ve ever been. And the US subprime market is simply not big enough to affect global credit markets in any substantial way.
At the margin, then, problems in the US subprime sector will surely have some effect on the US housing market — although, obviously, the effects will be greatest where the concentration of subprime loans is at its highest. But I still think that the big drivers of the US housing market in the recent present will continue to be the big drivers of the US housing market in the near future — which does give me some cause for optimism.
One thing is certain, however: foreclosures will continue to rise, which will be personally devastating to hundreds of thousands, if not millions, of homeowners. These personal tragedies will resonate on Capitol Hill, and the federal government might step in to help those individuals out — which might, in turn, help mitigate the economic effects of the foreclosures. What’s more, all the talk of hundreds of billions of dollars being “lost” in foreclosures is a mite misleading: all that equity has already been lost by the time the foreclosure proceedings begin.
The big losers will be the holders of the equity tranches of MBSs and subprime-backed CDOs, as well as those subprime originators who find themselves holding a bunch of scratch-and-dent loans they thought they’d sold already. So it makes perfect sense that subprime mortgage originators are closing shop. But I’m not shedding too many tears for them: they made a lot of money in the boom years, and then they relaxed their underwriting standards far too much at the end of 2005 and the beginning of 2006. They deserve to bear the consequences.
As for the holders of MBS and CDO equity, we’re talking very, very sophisticated investors and financial institutions here, who are almost without exception both willing and able to bear those losses. I see no systemic risks there.
Looking forwards, the number of subprime loans originated will go down, which means that issuance into the subprime MBS market will fall. That’s fine. As the yields on those subprime MBSs rise and the underwriting standards on them tighten, there will be enough demand for them to meet supply. So I’m not particularly worried there, either.
The one big event risk, which hasn’t happened yet, is a mass downgrade of MBSs carrying BBB and BBB— ratings, to junk status. If that happens, there will be forced selling of those MBSs by institutional investors who are not allowed to own speculative-grade debt, and spreads are likely to gap out substantially. Given how the markets managed to cope with the Ford and GM downgrades, however, I think they could probably cope with this as well. (Remember that Ford and GM bonds are unsecured, and therefore have much more room to fall.) What’s more, at this point there’s no indication that such a mass downgrade is going to happen.
Overall, then, I’m sanguine about the US mortgage and housing markets. There will be large losses, which will generally be sustained by entities well able to afford them. On an individual level, there will be some tragic losses of much-loved homes. But on a systemic level, I think there is relatively little to worry about.
Oh, and one last thing: there’s an argument that US GDP growth has been reasonably healthy of late mainly because of the amounts of equity that consumers are taking out of their homes and spending in shops. If they stop doing that, goes the argument, then the US faces recession — and house prices never do well in a recession. Again, my response is simple: show me, plausibly, the degree to which home equity withdrawal boosted GDP growth in recent years. Then we’ll have a good idea how much it could affect GDP growth on the downside.
I don’t deny for a minute that the number of defaults on subprime mortgages, especially subprime mortgages of the 2006 vintage, is eye-poppingly high, and rising.
Even this seems questionable to me. The numbers are high, but they appear to have always been high. Based on the latest MBA output, subprime looks stressed but not (yet) a disaster.
Thank you for yet another well written post! I’d be interested to know your take on Calculated Risks’s calculation of the impact of equity withdrawal on GDP? See —–
Good post. What is of interest to me is that the ‘housing bust’ may have 2 stages. This is the first. Prices will decline. Prices now reflect the too good to be true mortgages and those mortgages rely on increasing home values. The latter assumption is now false, so the too good to be true mortgages are going bad, and the component of housing prices which reflects this type of easy money will be the price decline.
But even after this, stage 2. Housing prices also reflect the other component of easy money: historically low long term interest rates. When this reverts to historic norms, housing prices will drop yet again. It’s not clear when stage 2 will happen, but I think the nervousness in the markets is related to a fear that stage 2 may be uncomfortably close.
My sense is that avoidance of a financial crisis requires that a good space of time separate stage 1 and stage 2.
Dissent — One reason why I am not quite as pessimistic as Roubini (Felix — you didn’t sway him: see is latest post, http://www.rgemonitor.com/blog/roubini, written in classic roubini style) is that 4.5% rates on the long bond should help put a floor under the market at some level, avoiding “stage 2.” At some point, everyone can just refinance into a fixed rate mortgage, and even if the rate isn’t quite as good as the teaser rate, it won’t be that bad ….
My sense is that to get stage 2 — and some mean reversion in long rates, which have been unusually low — central bank financing of the US has to end. And right now, best I can tell and contrary to my 04/05 priors, it is still rising …
I suggest you have a look at the Economist survey,”” on housing bubbles, published June 2005. It’s still germane because it talks about the generaly dynamics of housing manias, and their unwinding.
You are correct that subprimes by themselves shouldn’t cause this much damage. But it is calling attention to the fact that the supposed housing recovery is at the least oversold, and many believe conditions are actually getting worse, not better (see The Housing Bubble Blog, among other sources, which carries reports from all over the country).
Let’s look at it another way. Housing prices went up a ton with low interest rates. They haven’t corrected much with an increase in rates. And even before the increase in rates, the Economist estimated the overvaluation of housing in the US at 20%. That is a staggering number. And the Economist further observed that in corrections, housing prices often decline below “fair” value, in terms of their relationship to rents and wages.
As to the link to consumer spending, this was in the same article:
Another worrying lesson from abroad for America is that even a mere levelling-off of house prices can trigger a sharp slowdown in consumer spending. Take the Netherlands. In the late 1990s, the booming Dutch economy was heralded as a model of success. At the time, both house prices and household credit were rising at double-digit rates. The rate of Dutch house-price inflation then slowed from 20% in 2000 to nearly zero by 2003. This appeared to be the perfect soft landing: prices did not drop. Yet consumer spending declined in 2003, pushing the economy into recession, from which it has still not recovered. When house prices had been rising, borrowing against capital gains on homes to finance other spending had surged. Although house prices did not fall, this housing-equity withdrawal plunged after 2001, removing a powerful stimulus to spending.
I would point to the study published this week by Credit Suisse, which indicated subprime is not the only segment where lending standards evaporated. Credit quality is deteriorating across prime and non-prime. The alt-a stuff that’s so heavily used in California is also turning toxic but at a slower pace. (Not surprising because the teaser rates are longer for alt-a borrowers and shorter for subprime borrowers.) Having an 800 FICO score means nothing if you buy a house that’s 10x your annual income. This, I believe, is your real second wave.
Here, it has been proved with scientific precision that ARMs were the cause and not the effect of California’s run-up
Liquidity is declining. The tightened lending standards and the ever decreasing possibility to use a house to raise cash (prices are falling) is one factor. The JPY appreciation has unraveled part of the carry trade, reducing liquidity, and may continue to unravel. Not only in the US but other parts of the world as well. The CHF carry trade may begin to show surprises to US investors if the Fed, scared by declining economic activity indicators, were to cut rates. This will automatically increase the CHF value. Better said, the USD will devalue against other currencies. That will be another drag on liquidity. The Fed is fully aware of the falling liquidity and the possibility of a credit crunch. Today they injected an additional USD 20 billion to the primary dealers. That makes USD 57 billion these last two days. The subprime lenders’ problems, shop closing and defaults is another factor substracting liquidity from the market. The big money providers to this sector (Merrill Lynch, etc.) are now closing the spigot lest they get too involved in a crash which won’t bring them down, due to their size and capitalization, but which will hurt nevertheless. Merrill’s (MER) share price has already plunged from $97.53 on January 24 to $79.90 this Friday. So I wouldn’t be too sanguine. Just ask Brad Setser’s partner (NR) and I think he can give you much more sound arguments that I can.
The recent market downfall definitely is tied to the home mortgage bubble bursting…I just hope it doesn’t last too long.