Wednesday, September 23, 2015

Housing, A Series: Part 63 - More Evidence that Mortgage Repression is creating a bust, and there was no bubble.

An interesting article from Real Estate Economy Watch (HT: Todd Sullivan).  The gist:
We are deep into the best market for home sales in nearly a decade and the latest hard data shows that it is just as difficult to qualify for a purchase mortgage in July as it was last March–or even in March 2012.
Since the crisis, all-cash buyers, institutional buyers, and investors have pulled up some of the slack in home buying from households who have been pushed out of the credit market.  I think some of the headwinds in housing recovery and the continued excess returns to homeownership are due to organizational limits to how quickly those buyers can expand over large portions of a market that has always been dominated by owner-occupiers.

I have been waiting to see a recovery in mortgage credit markets as a signal that owner-occupiers might begin to return to the market.  There have been faint rays of hope, but generally this proposition has not yet come to pass.  There were signs late last year that regulatory constraints on conventional mortgage standards might loosen.  But, as the article suggests, this just may not be in the cards.  So, I think we will need to see some sort of substitution within the mortgage credit market.

This is why I would have hoped to see more expansion in closed-end real estate loans retained by the banks.  But, this has been disappointing.  There appears to be a burgeoning rent-to-own market, and new single family homes built for renting.  But, I think we also need to see growth in non-bank mortgage lending.  Private securitization markets will probably face many of the same headwinds as conventional securitization markets, so it looks like what needs to develop is non-bank, retained asset mortgage industry.  As with rented single family residence housing, this currently represents a small segment of the market, so, as with all of these residential real estate trends, it looks like there are organizational limits to the amount of growth that is possible in the short term, if it isn't going to come from traditional sources.  In this first graph, we can see that Ginnie Mae loan levels, after many years of stagnation, are the only source of mortgages that has grown since the crisis began, capturing back some of the low down payment market that had been moving into private pools.  All other mortgage holder groups have declined.  So, at least as of 1Q 2015, there had not been a resurgence of mortgage growth outside of federal agencies and banks.

We can see the effect on building by comparing residential and non-residential construction.  I think we can see several patterns here.  First, while non-residential construction has been generally declining as a portion of GDP for a long time, it has recovered to near pre-crisis levels.  (Note, incidentally, the very strong growth in manufacturing construction over the past year. In fact, it looks like manufacturing spending had previously begun to recover from a lull in the 2000s, but was interrupted by the recession.)  Residential construction spending rose to an unusually high level during the 2000s, briefly peaking at the start of 2006 about 50% above the normal level, then falling to about half the normal level from 2009 until today, with a slight upward trend.

I think residential construction spending was inflated in the 2000s, in part, because of the supply constrictions in the major cities, so that we were substituting billions of dollars of lumber and gypsum board in the exurbs for billions of dollars of valuable locations in the air above residential neighborhoods of our high density development-phobic cities.  This caused home sizes and construction spending to increase and productivity to decrease, relative to the alternative.  But, in any case, as with most indicators of residential investment, this appears to show a boom in residential construction in the 2000s followed by a larger bust in the 2010s.

But, was it a bubble, fueled by subprime mortgages and federal home ownership policies, that pushed home prices into unsustainable territory?

CoStar publishes indexes for Commercial Real Estate similar to the Case-Shiller Indexes for Residential Real Estate.  Here, I have graphed two CoStar indexes: Multi-Family Residential Housing and non-residential construction.  And I have compared them to the Case-Shiller National Home Price Index.  All are indexed to 100 at December 2000.

In terms of prices there is very little difference between the Case-Shiller index for single family homes and the CoStar index for multi-family properties.  And non-residential (retail, industrial, and office) prices follow fairly closely, generally rising and peaking about a year after residential prices, but peaking at the same level as residential prices, relative to prices in the 1990s.

After the crisis, single family home prices look like they were probably boosted by temporary homebuyer support programs in 2009 and 2010.  Multi-family and non-residential real estate bottomed in 2010 and have been rising by low double digits annually since then.  In the aggregate, both types of real estate are above pre-crisis price levels - multi-family real estate is much higher.  Meanwhile, single family home price increases are moderate.

Source : accuracy of real rate, from most to least is: green, red, purple.
Subprime loans and federal homeownership programs weren't pushing prices of commercial real estate up.  Given the parallel behavior of these real estate classes, there is no reason to believe they pushed up single family home prices either.  The main place on this graph where mortgage markets appear to have an effect is since 2011, where a lack of mortgage access is preventing single family home prices from recovering to normal levels.  (Here is a graph that shows the divergence after 2007 between real long term yields and implied yields on homes.)

Non-residential real estate has continued to rise, reflecting low real interest rates and general inflation over time.  Multi-family residential real estate prices are probably rising even faster than non-residential prices because the decade-long supply depression has created significant rent inflation and expected future rent inflation, and the foreclosure crisis pushed many families into rental housing, pushing up rents.  Also, multi-family housing, as a class, is especially exposed to the high rents from the dysfunctional cities.  Rents on single family homes are rising, too, but the difference is that multi-family housing owners have access to credit outside the single-family mortgage market.  So, those properties can be bid up to their reasonable market prices.

Single family homes may need to appreciate another 30% to trigger the complex set of market reactions that will lead to the robust new building that we desperately need.  Commercial real estate prices have risen that much more already because developers can usually charge market rents on commercial real estate in the big cities (Who is going to complain that rents on corporate offices are too high?), nobody frets about affordability indexes for corporations, and nobody demands that we fine or jail bankers for making "predatory" loans to commercial borrowers.  Single family homes face all of these obstacles.  In modern America, maybe the best you can hope for is that populists hate you enough that they stop coming up with imaginary problems that they have to fix for you.


  1. Excellent. I have long pointed out the mirror market in commercial real estate to residential.
    I wonder if the same building constraints, zoning, apply to commercial in the cities you identify is particularly bad in this regard.

    1. You know, it seems like there shouldn't be the same pressures on commercial rents. Commercial density is allowed to be so much higher in places like Manhattan than residential density. And telecommuting, e-commerce, etc. are reducing demand so that I see a lot of formerly commercial land that is now being in filled with residential. But the price behavior certainly suggests rents are similar, doesn't it? I suppose once a lot of land development decisions, etc. filter through the market, there should be enough substitution between commercial and residential development to cause expected net returns on capital, and therefore growth in rents, to be correlated over time.

  2. Office space we have. Retail space we have but I wonder if the retail space is where it is needed, close to dense populations.
    Build more housing.

    1. Yeah, commercial building is an interesting area. I'm not sure I have a solid understanding of the various factors involved. I agree with you that outlets for small scale activity have not been protected very well.

  3. Off topic... well, actually *on topic*... David Andolfatto has a post explaining how in the IS-LM model, present conditions could be explained by caps in the ability of investors to borrow:

    Kevin, this strikes me as an academic-macro slant on what you've been saying for weeks or months in this series, that frozen mortgage credit markets and an unwillingness in the regulatory apparatus to believe that higher home asset prices make sense has led to a capped demand for loanable funds, which brings us to zero interest rates and economic stagnation. The key text in the blog post that made me think of you is here:

    > The equilibrium interest rate is low--not because of Fed policy, but
    > because investment is constrained. Savers would love to extend more
    > credit if investors could be trusted and if regulatory hurdles were
    > removed. But alas, present circumstances do not permit this saving flow
    > to be released...

    David gets all this way without mentioning housing. I will suggest to him that he take a look at your stuff.


    1. Thanks Ken. I think he addressed housing in a recent post, but I think he assumes housing is operating in an equilibrium context, which is understandable. I would too, if I hadn't convinced myself otherwise over these last several months. I hope your suggestion to him garners some feedback. It would be really interesting to see what he has to say, though I'm afraid the sheer volume I have created on the topic is a kind of barrier to entry for new readers....

  4. bloomberg had an article the other day claiming that security guards, baristas, cashiers, and nursing aides r among the 20 job titles that have enjoyed the best wage gains this year. might the 2% wage growth across the economy b more consequential for consumption than it looks, if ever more of that gain is concentrated at the bottom among workers who consume the marginal dollar of income? what if there's not much slack. minimum wages r going up a lot in several major metros and seem to win everywhere on the ballot. could inflation rise in a step change when the cats who spend it all receive raises? this thought dissuades me from mREITs and p2p lending, for now

    1. I think the propensity of low income households to consume is an overestimated concept. I don't think it amounts to much. And, I think wage growth is more related to real production growth than it is to inflation.

      I think inflationary pressures from minimum wage hikes might come more from the damage it does to productivity than from the transfer to wage earners.

      Overwhelmingly, inflation now is coming from the national housing crisis due to the mortgage collapse. This actually makes consumption inflation inversely related to monetary and credit expansion. That should make the near future interesting....I just hope it doesn't become too interesting.

  5. "Overwhelmingly, inflation now is coming from the national housing crisis due to the mortgage collapse. This actually makes consumption inflation inversely related to monetary and credit expansion."

    This insight continuously blowing my mind about implications for pre-inflationary Fed headroom explains my reading here. it's like a Sherlock twist

    "I think inflationary pressures from minimum wage hikes might come more from the damage it does to productivity than from the transfer to wage earners."

    good point. still, if LFPR is not growing, while entry-level retail wages r growing along with retail productivity, this would seem to tempt workers in sectors with lower productivity growth, like child care and elder care, to sell cars or data plans. I'm not worried about CPI spooking the Fed so much as I'm worried about whether I should ship grandma to Thailand. I want to note that I'm not making a case for tightening, here. Granny's situation would not b much improved in a recession

    1. Did you see the paper that came out a few years ago about elder care? Apparently there is a mysterious countercyclical behavior in mortality. They found evidence that in recessions, nursing homes are better staffed, because, like you say, low income elder care workers find better opportunities during expansions.

    2. thanks for the tip; that's fascinating

      working in a 'home' during nursing school turned me against minimum wages, taxes on savings and work, stingy immigration controls, contemporary culture (low fertility, 'bowling alone'). even the nice places run by sincere, industrious people often feel like Guantanamo for the residents, for whom all attentions are rationed