The State of the Nation s Housing Eric S. Belsky Remodeling Futures Conference April 13, 21 www.jchs.harvard.edu
Existing Home Sales Improved then Retracted, While New Home Sales Are Still in the Basement Single Family Home Sales (Thousands, SAAR) 7,5 1,5 6, 1,2 4,5 9 3, 6 1,5 3 1963 1964 1965 1966 1967 1968 1969 197 1971 1972 1973 1974 1975 1976 1977 1978 1979 198 1981 1982 1983 1984 1985 1986 1987 1988 1989 199 1991 1992 1993 1994 1995 1996 1997 1998 1999 2 21 22 23 24 25 26 27 28 29 21 Existing Homes [Left Axis] New Homes [Right Axis] Sources: US Census Bureau; National Association of Realtors (R) through Feb 21
Home Building May Have Bottomed, But At Very Low Levels 2,5 Housing Construction Starts (Thousands, SAAR) 2,25 2, 1,75 1,5 1,25 1, 75 5 25 1959 196 1961 1962 1963 1964 1965 1966 1967 1968 1969 197 1971 1972 1973 1974 1975 1976 1977 1978 1979 198 1981 1982 1983 1984 1985 1986 1987 1988 1989 199 1991 1992 1993 1994 1995 1996 1997 1998 1999 2 21 22 23 24 25 26 27 28 29 21 Total Single Family Source: US Census Bureau through February 21
Only Single Family Starts Sustained Its Rebound into 21 Indexed Quarterly Value (Jan 29 = 1) 21 19 17 15 13 11 9 Q1_28 Q2_28 Q3_28 Q4_28 Q1_29 Q2_29 Q3_29 Q4_29 Feb 21 SF Starts New SF Sales Existing SF Home Sales Sources: US Census Bureau; National Association of Realtors.
Home Prices May Be Stabilizing 175 Real Single Family Home Prices (Indexed Jan 2 =1) 15 125 1 75 5 1968 1969 197 1971 1972 1973 1974 1975 1976 1977 1978 1979 198 1981 1982 1983 1984 1985 1986 1987 1988 1989 199 1991 1992 1993 1994 1995 1996 1997 1998 1999 2 21 22 23 24 25 26 27 28 29 S&P Case Shiller HPI NAR Median Home Price FHFA (formerly OFHEO) Note: All prices are seasonally adjusted and inflation adjusted using the CPI-U for All Items. Sources: S&P/Case-Shiller 2-city composite house price index (SA); National Association of Realtors (R), median single family home price (SA, in thousands) ; Federal Housing Finance Agency, purchase-only house price index (SA).
Notes: Monthly mortgage costs assume a 1% downpayment and Freddie Mac conventional 3-year fixed interest rates, and are adjusted for inflation by the CPI-U for All Items. Metro areas evaluated are the 79 metros in the National Association of Realtors series for which there are data from every quarter 1989-2 and 29:3. Sources: JCHS calculations using the National Association of Realtors, Median Existing Single-Family House Price and Moody's Economy.com, Average Household Income. Housing Affordability Is the Biggest Plus - And Interest Rates Matter Share of Metros (Percent) 1 9 8 7 6 5 4 3 2 1 Real Monthly Mortgage Costs Ratio of Median Price to Average Income 29:4 Level Relative to 1989-2 Average: More than 1.% Below -1. to - 5.% Below 5. Below to 5.% Above 5. to 2.% Above More than 2.% Above
The Road to Recovery www.jchs.harvard.edu
Home Building Recoveries: The Top Line Takes sustainable turnaround in new home sales to bring single-family starts back (it s the demand) New home sales are the Punxsutawney Phil of housing and economic recoveries Employment growth lags new home sale turnarounds, typically by two quarters Only 1 recovery of last 4 failed after GDP turned around coming out of a recession It took a Fed-engineered spike in interest rates to derail the recovery, small increases in rates or home prices have not
Lasting Turnarounds in New Sales Started Just Before Sustained Turnarounds in GDP Growth 14 New Single Family Home Sales (s, SAAR) 12 1 8 6 4 2 1971Q1 1972Q1 1973Q1 1974Q1 1975Q1 1976Q1 1977Q1 1978Q1 1979Q1 198Q1 1981Q1 1982Q1 1983Q1 1984Q1 1985Q1 1986Q1 1987Q1 1988Q1 1989Q1 199Q1 1991Q1 1992Q1 1993Q1 1994Q1 1995Q1 1996Q1 1997Q1 1998Q1 1999Q1 2Q1 21Q1 22Q1 23Q1 24Q1 25Q1 26Q1 27Q1 28Q1 29Q1 New Home Sales True Trough Possible False Bottom Note: Lightly shaded areas mark official recessions. Darkly shaded areas mark beginning of sustained recoveries in real GDP at ends of recessions. Sources: US Census Bureau; US Bureau of Economic Analysis
Signs of Lasting Turnarounds in New Home Sales Before Sales Bottom: Nominal mortgage rates were falling Real monthly payments were falling Inventory of new homes for sale were in decline several quarters After Sales Bottom: X??? Real GDP turned positive two quarters after new sales trough Total employment hit bottom two quarters after new sales trough GDP growth remained positive for next 8+ quarters of sales recovery Once employment turned, stayed positive at least 5 more quarters Inventory of new homes bottomed 1-6 quarters after turnaround
Past Rebounds in New Home Sales Lasted 4 Quarters Before Any Noticeable Declines, Failed Recovery Did Not Lasting Recoveries Failed Recovery 8 75 7 65 New Home Sales (s, SAAR) current rebound to-date 8 75 7 65 New Home Sales (s, SAAR) current rebound to-date 6 6 55 55 5 5 45 45 4 35? 4 35? 3-4 -3-2 -1 +1 +2 +3 +4 +5 +6 +7 +8 Quarters Trough in Quarters After Trough Before New Home Sales Mid 197s Recovery Second Double Dip Recovery Early 199s Recovery Current 3-4 -3-2 -1 +1 +2 +3 +4 +5 +6 +7 +8 Quarters Trough in Quarters After Trough Before New Home Sales First Double Dip False Rebound Current Source: JCHS tabulations of US Census Bureau data
Strength of Employment Growth Has Mattered Most to The Strength and Sustainability of Past Starts Recoveries Growth in Total Starts Through First 8 Quarters after Trough in New Home Sales (Percent) 1 8 6 4 2-2 Second Double Dip Mid 7s Early 199s First Double Dip Growth in Employment Through First 8 Quarters after Trough in New Home Sales (Percent) 6 5 4 3 2 1 Second Double Dip Mid 7s Early 199s First Double Dip
Employment Losses This Recovery Are Staying Worse Longer +1.5 Employment Growth (Percent) +1. +.5. -.5-1. -1.5-2. +1 +2 +3 Quarter After Trough in New Home Sales Mid 197s Recovery Early9s Recovery Recovery After Second Double Dip Current Note: Quarterly Employment Growth Rates Around Troughs in New Home Sales Source: JCHS Tabulations of US Census and BLS data
Falling Mortgage Interest Rates Have Given Less Lift to Sales Following the Current Trough +3. +2. +1.. -1. -2. -3. -4. -5. Indexed Mortgage Rate Relative to Rate at Trough (Percentage Point) +1 +2 +3 Quarter After Trough in New Home Sales Mid 197s Recovery Early9s Recovery Recovery After Second Double Dip Current Source: JCHS Tabulations of Freddie Mac 3 Year Fixed Conventional Conforming Mortgage Rates
But Nominal Rates This Time Are At Very Low Levels 18 Nominal Mortgage Interest Rate (Percent) 16 14 12 1 8 6 4 2 +1 +2 +3 Quarter After Trough in New Home Sales Recovery After 8s 2nd Dip Early9s Recovery Mid 197s Recovery Current Source: JCHS Tabulations of Freddie Mac 3 Year Fixed Conventional Conforming Mortgage Rates
Among Metros with Large Cyclical Permit Declines, It Took 3 Years Before Worst Decliners Saw Stronger Rebounds Change in Annual Single Family Permits Among 5 Largest Markets with Declines >1% (Percent) 1 8 Note 15-2% Average Change 6 4 2-2 -4-6 -8 Median Peak to Trough Decline, 198s to Early 199s After 1 Year After 2 Years After 3 years After 4 Years Median Recovery Ten Worst Declines in Permitting All Other Metros Ten Smallest Declines in Permitting Note: Includes the top 5 homebuilding markets in that cycle, as ranked by permits issued in 1986. But excludes four markets that had insignificant (less than 1%) cyclical declines. Source: JCHS Tabulations of US Census Data
Conclusions Key indicators of real turnaround in new home sales in place except for timing of employment rebound Strength of the recovery is likely to depend on the timing and strength of job growth key to watch The size of metro permit rebounds in the first years of last recovery were not related to size of prior declines The headwinds this time are strong and without precedent high foreclosures/tightened standards Last quarter s decline in new home sales bears noting History is an imperfect guide, but it s all we get Consensus forecasts by others call for about 7,- 75, starts in 21 and 1.-1.15 million in 211
45 4 35 3 25 2 15 1 5 The Pipeline of Serious Delinquencies Is Still Rising Share of Loans 9+ Days Delinquent or In Foreclosure (%) Q1 26 Q2 Q3 Q4 Q1 27 Adjustable Subprime (NSA) Adjustable FHA Adjustable Prime Q2 Q3 Q4 Q1 28 Q2 Q3 Q4 Q1 29 Fixed Subprime Fixed FHA Fixed Prime Q2 Q3 Q4 42.7 22.1 18.1 12. 9.2 5. Source: Mortgage Bankers Association
Note: Good Loans Defined as all other loans not having high LTV & low FICO nor Low LTV & High FICO. Source: Barclays Capital. GSEs: Back to the Future, US Interest Rates Strategy, 11 December 29. Lending Standards Even in the Prime Market Have Tightened Dramatically 1 9 8 7 6 5 4 3 2 1 Share of GSE Originations (Percent) 26 29 "Better": FICO >75 & Original LTV <75 "Good" "Worse": FICO <69 & Original LTV >85
Nominal Annual Price Declines Have Been More Severe and Widespread than Any in Recent History 7 Share of the 73 Metros Covered by CMHPI since 1975 6 5 4 3 2 1 1981 1982 1983 1993 1994 1995 27 28 29 Less than 1 Percent 1 to 5 Percent Greater Than 5 Percent Note: Metros included are the 73 Metros fully covered by CMHPI data since 1975. Annual values are averages of quarters. Source: JCHS tabulations of Freddie Mac CMHPI data.
Estimating Long-Run Demand Estimates are of long-run sustainable demand not a prediction of actual completions and placements Actual construction influenced by under or over supply entering and exiting the period Household growth projected under two immigration assumptions and with headship rates held constant Two methods used for vacancies and net removals Each projected separately and conservatively Using average ratio of completions plus placements to household growth for ten-year periods ending 1994-1998 (which contains the 87-91 housing cycle)
Recently Headship Rates for Younger Adults have Declined Slightly and Older Adults Have Increased Slightly, but.7.6.5.4.3.2.1 Headship Rate 15-19 2-24 25-29 3-34 35-39 4-44 45-49 5-54 55-59 6-64 65-69 7-74 75+ Total Age of Household Head 23 CPS 28 CPS Source: Masnick, George S. and Eric S. Belsky. 29. Household Projections in Retrospect and Prospect: Lessons Learned and Applied to New 25-225 Projections. JCHS Working Harvard Paper W9-5. University
Immigration Has Become an Increasingly Important Driver of Household Growth 45% 4% 35% 3% 25% 2% 15% 1% 5% Foreign Born's Share in Total Household Growth 15.8% 29.8% 4.5% % 198s 199s 2-25 Source: JCHS tabulations of the 198, 199 and 2 decennial Censes, and the 25 American Community Survey.
Household Growth Is On A Higher Path But Immigration Is The Key Wildcard 1.6 1.5 Average Annual Household Growth (Millions) 1.48 1.4 1.3 1.25 1.25 1.2 1.15 1.1 1. 1995-2 2-28 21-22 21-22 Assuming Census Projections of Immigration Assuming Half of Census Projections of Immigration Notes: To adjust for rebenchmarking, household growth in 22-23 is assumed to be the same as the average annual growth in 2-25. JCHS projections using 28 US Census Bureau Population projections. Sources: US Census Bureau, Housing Vacancy Survey; JCHS Working W9-5.
Summary of Findings Projection of separate elements yields 21-22 estimates of between 17.2 (low immigration) and 19.7 million (high immigration) The historic ratio of completions and placements to household growth yields 21-22 estimates of between 17.1 and 2.2 million Important caveats: Projections of demand for single and multifamily homes completed + manufactured homes placed Assumes enter and exit the period in market balance Odds of being in perfect balance in both periods low No ironclad measures of when markets are in balance Think of as underlying current of demand
Impact of Entering Oversupply Assumptions on Projections for 21-22 (With Markets Assumed to be in Equilibrium by 22) Assumed Oversupply Entering 21 Baseline Average Annual Completions & Placements 21-22 needed to maintain long-run average 1.72-1.97 Million 5 Thousand 1.67 1.92 Million 1. Million 1.62 1.87 Million 1.5 Million 1.57 1.82 Million Notes: Ranges of annual average completions needed to maintain long-run average 21-2 determined by applying both JCHS 29 low and high household projections. Sources: US Census Bureau, JCHS Household Projections, JCHS Working Paper W9-5; JCHS Working paper W7-7.
Had Actual Demand Held Up to Long-Run Levels, Oversupplies Entering 26 Would Have Been Worked Off 2,25 2, 1,75 1,5 1,25 1, 75 5 25-25 -5-75 -1, -1,25-1,5 Finished Units In Excess of Long Run Demand ('s) Finished Units in Excess of Long-Run Demand (Thousands) 1.2 1.8.6.4.2 Entering 26 Entering 29 Entering 21 Completions & Manufactured Home Placements Accumulated Excess Units- Conservative Demand Estimate Accumulated Excess Units- Aggressive Demand Estimate Entering 26 Entering 29 Entering 21 Completions & Manufactured Home Placements Accumulated Excess Units - Conservative Demand Estimate Accumulated Excess Units - Agressive Demand Estimate Notes: Finished Units are the sum of housing completions and manufactured home placements. Sources: JCHS Calculations of data from the US Census and JCHS Research Note N9-1.
Owner Improvement Growth Exceeded Expectations 2-26 But Now Below Trend Line Owner Improvements (Billions 28$) 35 3 25 2 15 1 5 1995 1996 1997 1998 1999 2 21 22 23 24 25 26 27 28 29 21 211 212 213 214 215 Owner Improvements Trendline Growth Note: Trendline growth calculated by applying low-immigration series 29 JCHS household growth projections by age, hhtype and minority status to 1995-2 percentage growth in AHS average owner investment within those categories. Sources: JCHS tabulations of data from the US Census Bureau, American Housing Survey, and JCHS Household Projections; JCHS Working Paper W7-6.