A New Leading Indicator for Housing

1
A NewLeading Indicator for Housing

• Dr. David Lereah
• Chief Economist
• NATIONAL ASSOCIATION OF REALTORS

New York City, March 30, 2005
2
Boom in Existing-Home Sales
Thousand units
3
New-Home Sales
Thousand units
4
Single-Family Housing Starts
Thousand units
5
Strong Demand But Not Enough Inventory
6
Home Prices are Getting Bid Up
change from a year ago
7
Economic Outlook
8
Housing Bubble?
9
Recent Price-Surging Markets
10
Past Bubbles PopDue to Combined Job Losses and
Rampant Oversupply
11
Payroll Jobs on the Rise - Supports Housing Demand
2.4 million net new jobs in 12 months
12
Not Enough HomesPast 10 years (1993 to 2003)
Mismatch
NAHB estimates
13
Next 10 Years
Unlikely to be oversupply
NAHB estimates
14
Housing Outlook
15
Pending Home Sales Index (PHSI)

• A New Leading Indicator For The Housing Sector

16
PHSIWhy A New Index?

• Housing sector is vital to the U.S. economy
  contributes 20 to GDP
• EHS is a major economic indicator, closely
  watched by the Housing Industry , Financial
  Analysts, and the Federal Reserve
• EHS reports recent history
• PHSI will provide direction of future housing and
  economic activity

17
PHSI

• Few housing indicators are considered leading
  indicators of EHS purchase applications,
  new-home sales, housing starts
• But the relationship between these indicators and
  EHS is less than satisfying
• To better predict existing-home sales we have
  created a new leading indicator- the PHS Index

18
(No Transcript)
19
PHSI
We plot the National PHSI lagged 2- months vs.
unadjusted EHS We observe a strong statistical
relationship (Correlations .87)
20
PHS Background

• When a seller accepts a contract it is recorded
  in a Multiple Listing Service (MLS) as a pending
  home sale
• PHSs are recorded for virtually all MLSs across
  the country
• NAR began collecting PHS data last year, asking
  MLSs for data back 3 years
• Compiled a sample of over 100 large MLSs
• We assessed the relationship between PHS EHS

21
PHS DataLead-time analysis
Pending sales correlate with closed sales The
lead-time varies by local area Strongest
relationship is with PHS leading closed sales by
two months (correlation .90)
22
PHS DataLead-time analysis
23
PHS DataLead-time analysis

• 38.2 closed in 1 month, 42.7 closed in 2 months
• Nearly 10 had pending sales close 3 months later

24
PHS DataLead-time analysis

Correlations stronger at local level PHS Index
will be even more accurate at the local
level (correlation .94)
25
PHSIMethodology

• Benchmark
• From 100 MLS sample we derive national/regional
  benchmark indices
• Starting year is 2001 first year with largest
  and most complete sample
• Raw data create regional indices
• Weighted average of regional indices used to
  arrive at national PHS index
• Percent changes in raw data used to calculate
  indices for 2002 2003
• Methodology presented received input from
  Housing Analysts and Federal Reserve

26
PHSIMethodology

• Data from 100 MLSs are used to create the
  historical series (2001 to 2003)
• Not all MLS report this data to us on a monthly
  basis
• We create a separate survey of large brokers to
  supplement our sample of MLSs
• Combining MLS data with broker data gives us a
  monthly sample size covering 50 of EHS sample

27
PHSIMethodology

• MLS data is preferred -- large broker data used
  to fill in gaps, duplicates are rejected
• Sample is structured to mirror EHS sample
• We note that not all pending sales get sold. We
  monitor the fallout rate based on a sub-sample of
  MLSs survey of practitioners
• When data is available, we will construct
  National/Regional fallout rates

28
PHSI Methodology
29
PHSI

• EHS bo b1 (PHSlag1) b2 (PHSlag2) e

Non-seasonally Adjusted
Predicted vs Actual
Actual
Predicted
Absolute
value
value
Error
Chg
October
549,000
542,000
7,000
-1.3
539,000
November
530,000
9,000
1.7
Converted to
Seasonally Adjusted
Actual
Predicted
Predicted vs Actual
Absolute
value
Error
value
Chg
October
6,750,000
6,701,000
49,000
-0.7
November
6,940,000
7,061,000
121,000
1.7
30
PHSI

• Since pending home sales measure actual
  existing-home sales, the PHSI will provide an
  accurate and reliable indicator of future home
  sales activity
• It has the potential to become the most relevant
  and important indicator in the housing market
• Upon more comprehensive sampling, NAR will look
  to report pending sales information for state
  local markets

31
PHSIMonthly Table
32
PHSIAnalyst Applications

• Valuing mortgage-backed securities
• PHS to assess demand in a local area evaluate
  perform risk analysis mortgage pools
• Real estate-related company valuations
• PHS provides geographic risk exposure to
  evaluate stocks (Home Depot, Lowes, etc)
• Market analysis implications
• PHS used to assess economic conditions and
  direction of financial markets

33
PHSIAnalyst Applications

• Regional/Local business risks
• A sharp decline in PHS might be first indication
  of home price declines local market weakness
• Local housing trends useful in assessing
  commercial and multifamily markets
• Migration patterns population growth
• Identify saturated markets or high growth areas
• Other

34
PHSIDemand from Housing Industry

• Home Builders
• Mortgage Bankers/Lenders
• Real Estate Brokerage Firms Agents
• Settlement Service Providers
• Relocation Companies
• Commercial/Residential Real Estate Investors
• Other

35
A NewLeading Indicator for Housing

• Dr. David Lereah
• Chief Economist
• NATIONAL ASSOCIATION OF REALTORS

New York City, March 30, 2005