US deleveraging, 3 – housing pressures

We will attempt to knit up the series on deleveraging in this post, as we suspect that those not inclined to advocate for the ongoing observation of debt flows believe that those who are so inclined might own an Irving Fisher t-shirt or two. That’s not (quite) the case here, but no apologies will be made for our long forays into the well-tended gardens of the Financial Accounts of the US. For it is there, along with jobs data by industry from the Bureau of Labor Statistics, that we believe we can see the outlines of a broad post-panic US economic policy by design, where monetary, banking, credit, and fiscal policies, as well as legislative activities and interpretation, have all strongly impacted the landscape.

We return to where we started this series of posts: residential real estate, where residual drag remaining from deleveraging can be found at the time of writing, July 2016. Why does it matter? Isn’t it proper, and just, that residential real estate has endured such a lengthy and deep adjustment? It is our view that the picture painted by national data show undue repression of the residential real estate sector, which hasn’t quite yet ended after a decade since the market peaked. The cost is not only in terms of lost GDP, but in terms of distributional equity as well. Not a good direction for policy given trends in income for the last twenty years or so.

The distortions in resource allocation caused by market repressions cannot be completely foreseen in advance. However, given the patterns in mortgage underwriting during the boom and bust stages, the limited assistance to underwater homeowners, the slow and unsteady advance in job creation, and subdued wage growth, all of that on top of deleveraging by financial firms and state and local governments, it is unsurprising that the cost, availability and quality of housing are proving unfriendly toward a swath of society, particularly young adults and low-income persons. This despite notable private investment flows into rental housing, seen in multi-family housing starts and the acquisition of foreclosed homes to let.

Graph 1 shows the annual change in Personal Consumption Expenditure (PCE) price indexes for housing from 2002 to 2014, covering the last two expansions. Unfortunately, the scope of this post precludes discussion of the imputed rent method of calculating owners’ housing expenditures, and the differences between the PCE and CPI approaches to measuring the cost of housing.

Graph 1

Source: US Department of Commerce, Bureau of Economic Analysis, Personal Income and Outlays, Table 2.5.4. Price Indexes for Personal Consumption Expenditures by Function. http://bea.gov/iTable/iTable.cfm?ReqID=9&step=1#reqid=9&step=1&isuri=1
Source: US Department of Commerce, Bureau of Economic Analysis, Personal Income and Outlays, Table 2.5.4. Price Indexes for Personal Consumption Expenditures by Function. http://bea.gov/iTable/iTable.cfm?ReqID=9&step=1#reqid=9&step=1&isuri=1

Note that of the 13 years represented, the rental deflator exceeded two percent for all but two of those years, and increased a total of 37 percent over that time, versus 27 percent for the full PCE price index. Graph 2 shows the PCE deflator for rental housing against indexed real median household incomes, over a longer time span.

Graph 2

Source: US Department of Commerce, Bureau of Economic Analysis, Personal Income and Outlays, Table 2.5.4. Price Indexes for Personal Consumption Expenditures by Function. http://bea.gov/iTable/iTable.cfm?ReqID=9&step=1#reqid=9&step=1&isuri=1; US. Bureau of the Census, Real Median Household Income in the United States [MEHOINUSA672N], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/MEHOINUSA672N
Source: US Department of Commerce, Bureau of Economic Analysis, Personal Income and Outlays, Table 2.5.4. Price Indexes for Personal Consumption Expenditures by Function. http://bea.gov/iTable/iTable.cfm?ReqID=9&step=1#reqid=9&step=1&isuri=1; US. Bureau of the Census, Real Median Household Income in the United States [MEHOINUSA672N], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/MEHOINUSA672N
The State of the Nation’s Housing 2016 report, prepared by the Joint Center for Housing Studies of Harvard University (JCHS 2016), is studded with worrisome data on US households and housing. The report (p. 15) points out that households earning under USD 25,000 per year account for 44 percent of the net growth in households from 2005 to 2015. (The change in the number of households during that period is dominated by 6.4 million nonfamily households, 2015 average size 1.25 persons, compared to 4.9 million family households, average size 3.21 persons. Cold comfort. Source: Bureau of the Census, Current Population Survey, Series HH-1.)

The following bulleted observations come from the JCHS report, unless otherwise noted.

  • The number of renters increased by over 21 percent from 2007 to 2015, to a total of 42.6 million. The number of home owners fell over that period by 0.6 percent (source: Bureau of the Census, Current Population Survey, Series H-111, Table 7).
  • The number of cost-burdened renters, defined as those paying more than 30 percent of income for housing, is put at 21.3 million, half of all renters. The number of severely-burdened renters, paying more than 50 percent of income for housing, represent over a quarter of all renters.
  • Only one in four low-income renters eligible for housing assistance receives it. Units affordable to low-income renters are often occupied by higher-income renters. The National Low Income Housing Coalition estimated that there were 57 affordable and available units for every 100 very low-income renters in 2014.
  • Homeownership is a vehicle for building wealth that is lost to the permanent renter, and an investment that, unlike securities, provides housing services. The JCHS report, which describes a “close link” between homeownership and wealth, notes that over 75 percent of households with net worth less than USD 25,000 are renters. The Federal Reserve’s Survey of Consumer Finances (SCF) puts 2013 median net worth for homeowners at USD 195,400; for renters or others, USD 5,400. However, it should be kept in mind that many renters are young and building wealth. More detailed data is needed. It is safe to say that the forced saving rendered through the vehicle of the amortizing mortgage is powerful, especially in a low-interest rate environment. The amortizing mortgage is one of the remarkable developments in US finance in the wake of the Great Depression.
  • The racial divide in homeownership appears to support the connection between homeownership and wealth. 2013 median net worth for whites is USD 142,000 (source: SCF), with a homeownership rate of 73 percent (source: Bureau of the Census, 2013 American Housing Survey). 2013 median net worth for nonwhites is USD 18,100 — a thoroughly disturbing figure — and homeownership rate 47 percent. It has been well-noted elsewhere that from the 2004 peak through 2015, blacks lost more ground in homeownership than whites and other minorities, 6.7 percentage points, versus 4.1 percentage points for whites. The SCF shows a considerably greater decline in median wealth for nonwhites — 36 percent — compared to whites — 17 percent — over the period 2007 to 2013.

There is more. There is the threat of eviction for burdened dwellers, and the expenses borne if eviction comes to pass. There is the scourge of homelessness, with their numbers estimated at roughly 565,000; 75,000 in New York City alone (JCHS 2016). There is the cost trade-off to take on the hardship of distant residential location, which can be made even more difficult by heavy traffic and poor public transit options. Foreclosures are still running at twice the rate as they were prior to the crisis (JCHS 2016). CoreLogic estimated that 9.4 million homes had been forfeited through foreclosure, short sales, and deeds-in-lieu of foreclosure from 2007 through 2015. To get some sense of scale, that represents 12.5 percent of the number of owner-occupied units in 2007.

There is the difficulty for would-be home buyers to generate a 20 percent down payment — as of April 2016, USD 46,740 on the US median-priced existing home as cited by the National Association of Realtors (NAR) — with student loans, pressured incomes, and underwater home prices throwing grit into those wheels of commerce. The NAR calculates that the income required to afford payments on a median-priced existing single family home is USD 42,288 (as of April 2016; based on a 25 percent ratio for housing expense to income and a 20 percent down payment), which seems reasonably accommodating: 2014 US median household income is USD 53,657 in 2014 CPI-U-RS adjusted dollars (source: US Census Bureau, Income and Poverty in the US 2014). Note that there is significant regional variation, with qualifying income in the Midwest of USD 33,840, and the West of USD 60,672.

Pressures in housing at all levels could be relieved were it not for the suppression caused by policy changes, including the barely roused state of the private mortgage-backed securities (MBS) market. To see that home mortgage debt outstanding in 2015 expanded by a paltry USD 138 billion saar off a base of USD 9,404 billion is frankly, shocking. To put that in perspective, prior to the most recent recession, one has to go back to 1984 to find a lower origination figure (USD 127.6 billion, nominal), and recall that there was a significant mortgage overreach in the late 1980s. Other consumer credit (primarily student loans, autos, and credit cards) increased by USD 232 billion in 2015; growth in the consumer credit segment similarly exceeded home mortgage borrowing in Q1 2016.

Of the USD 152.2 billion originated in mortgages on 1–4 family properties in 2015 (Federal Reserve Board Financial Accounts of the US, Table F.218, June 9, 2016 release; includes USD 14.1 billion borrowed by nonfinancial businesses): originations of US-chartered depository institutions account for USD 55 billion (!); credit unions, USD 33 billion; GSEs, along with agency- and GSE-backed MBS, USD 140 billion. Private label MBS paydowns totaled USD 73 billion, not to be sniffed at. Finance company outstandings also shrank, by USD 19 billion.

US-chartered depository institutions returned to net origination of mortgages in 2014, last experienced in 2007. GSEs and agency- and GSE-backed MBS remained net originators through the cycle. Remember the gangplank walk that the agencies were forced to take in 2008? Some of you may recall that a FNMA/FHLMC untethering was positioned in July of that year, and the market instability that talk fomented, culminating in conservatorship for the agencies in September 2008. But, yet, they move, the GSEs, while banks,  and the private MBS market, look bound and gagged.

Is this a desirable state of the world, policy-wise, at this point in the cycle? Was it necessary to spend nearly a decade to rehabilitate the mortgage industry? – not that we are claiming that the rehabilitation is complete. Are we to believe that the stasis in the private MBS market is acceptable? It has been eleven years since housing peaked in the US, and since then waves of young adults entering or trying to enter the work force have been handed a package of economic disappointments. Why should it take so long for a steady state to re-emerge? Is there any accountability for the delay? Why isn’t a steadier provision of credit throughout the cycle made a policy goal?

References

JCHS. 2016. The state of the nation’s housing 2016. Joint Center for Housing Studies of Harvard University.                           http://www.jchs.harvard.edu/research/state_nations_housing