Supply Elasticity of Houses in Regional NSW Xiang Ling Liu & Glenn Otto UNSW EMG Workshop 5 November 2014
Perception that Housing Supply in Australia is Inelastic there have been a number of factors on the supply side that have combined to keep the supply of new housing below where it would have been in a more responsive environment. As a result, we have had the combination of higher prices and lower supply than might otherwise have occurred. (Tony Richards, 2009, RBA) 2
What is the supply elasticity of residential housing in Australia? All Housing Ball, Meen and Nygaard (2010) Australia 0.55 Gitelman and Otto (2012) Sydney 0.33 Liu and Otto (2014) Sydney 0.51 Houses Apartments Gitelman and Otto (2012) 0.18 0.53 Liu and Otto (2014) 0.22 0.80 3
Why Regional NSW? Relatively good disaggregated data on house prices Is housing supply more elastic in regional NSW than in Sydney? 4
Data Cross-section unit: Local Government Area (LGA) 101 LGAs Prices Median sales price for non-strata dwellings Raw data are quarterly observations beginning in 1991:1 Missing observations for some LGAs are guesstimated Converted to annual frequency by averaging quarterly calendar year obs Sample 1991 2012 All series converted to real house prices using Sydney CPI 5
Thousands ($) Real House Prices in LGAs of Griffith and Nambucca 350 300 250 200 150 100 50 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Griffith Nambucca 6
House Stocks Number of private non-strata dwellings Census data provide estimates for 1991, 1996, 2001, 2006 and 2011 Inter-censual years are interpolated using data on building approvals for LGAs No accounting for quality of housing In most regional LGAs, non-strata dwelling = detached house 7
Number of Houses Number of Non-Strata Dwellings in LGAs of Griffith and Nambucca 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Griffith Nambucca 8
Regional Groupings for LGAs 11 regions (Number of LGAs) Coastal Inland Hunter (11) Central West (13) Illawarra (5) Far West (1) Mid-North Coast (7) Murrumbidgee (12) Richmond-Tweed (6) Murray (9) South-Eastern (14) North-Western (11) Northern (12) Coastal means that at least some LGAs have a coastline. 9
Regional Map 10
% change Growth of Housing Stock and Real Prices by Region, 1991-2011 140.0 120.0 100.0 80.0 60.0 40.0 20.0 0.0-20.0 Richmond-Tweed Mid-North Coast Hunter Illawarra South-Eastern Northern Central West North-Western Murrumbidgee Murray Far West Non-Strata Houses Real Prices 11
Increase in House Prices: 2002-2004 Feature of the data is a large increase in real house prices in most LGAs from 2002-2004 12
Normalised Real House Prices for Richmond-Tweed 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 1985 1990 1995 2000 2005 2010 2015 13
Normalised Real House Prices for North Western Region 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 1985 1990 1995 2000 2005 2010 2015 14
NSW-Wide Common Shock Increase in demand curve for houses Reduction in supply curve for houses If we assume the former, then can use the increase in house prices over 2002-2004 to estimate supply elasticity: % Δ in stock of houses 2001 to 2005 % Δ in real house prices 2001 to 2005 15
Elasticity Naïve Estimate of Supply Elasticity by Region 0.12 0.10 0.08 0.06 0.04 0.02 0.00 CW FW HT IW MN MB MR NW NT RT SE -0.02-0.04 16
Formal Supply Elasticity Estimates Use annual time-series data from 1991-2012 to estimate supply elasticity for each of 101 regional LGAs 17
(Very) Simple Model Supply Curve Reduced Form lnh t i = α i + β i lnp t i + u t i lnp t i = π 0i + π 1i lny t i + π 2i lnn t i + π 3i R t + v t i H = stock of non-strata properties P = real median price of non-strata properties Instruments Y = real income per taxpayer N = resident population R = real 10 year bond rate 18
F-stat Instrument Quality (income and real rate) 350 300 250 200 150 100 50 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 LGA 19
Density Functions for Regional and Metropolitan Supply Elasticity 5 4 3 2 1 0-0.2 0 0.2 0.4 0.6 0.8 1 1.2-1 Supply Elasticity Regional Houses Sydney Houses 20
Log Houses Estimating Supply Elasticity: Nambucca (IV Estimate = 0.41) 9.05 9 8.95 8.9 8.85 8.8 8.75 y = 0.3106x + 7.1505 R² = 0.6852 8.7 8.65 8.6 5 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 Log Price 21
Log Price Estimating Supply Elasticity: Nambucca (1/2.21 = 0.45) 5.9 5.8 5.7 5.6 5.5 5.4 y = 2.2064x - 14.07 R² = 0.6852 5.3 5.2 5.1 5 4.9 8.6 8.65 8.7 8.75 8.8 8.85 8.9 8.95 9 9.05 Log Houses 22
Another Estimator ARDL Bounds Procedure (Pesaran, Shin and Smith, 2001) lnh i i t = δ H0 i i + δ HP lnp t 1 i i + δ HH lnh t 1 i + ω HP ln P i i t + v t i H 0 : δ HP i = δ HH = 0 Test for levels relationship between lnh and lnp If F-stat is sufficiently small then don t reject null If F-stat is sufficiently large then reject null Range for F-stat where you need to pre-test for variables being I(1) or I(0) If null is rejected, then estimate long-run supply elasticity by δ HP i δ i HH 23
IV Estimates Correlation between IV and ARDL Estimates 1 0.8 0.6 0.4 0.2 0-0.6-0.4-0.2 0 0.2 0.4 0.6 0.8 1 1.2-0.2 ARDL Estimates 24
Growth in Real House Prices and Supply Elasticity As a result, we have had the combination of higher prices and lower supply than might otherwise have occurred. (Tony Richards) Do LGAs with lower supply elasticity have higher average capital gains? 25
Real Capital Gain (% pa) Supply Elasticity and Real Capital Gains: Coastal Regions 9.0 8.0 y = -2.0525x + 4.3696 R² = 0.0771 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Supply Elasticity 26
Real Capital Gain (% pa) Supply Elasticity and Real Capital Gains: Inland Regions 8 7 6 5 y = -0.9877x + 2.7978 R² = 0.0206 4 3 2 1 0-0.2 0 0.2 0.4 0.6 0.8 1-1 -2 Supply Elasticity 27
Why is Housing Supply in Regional NSW so Inelastic? 2 observations Average supply elasticity for houses in regional NSW (0.32) is not much greater than in Sydney (0.2) There are no regional LGAs with elastic housing supply 5 largest: Murray (0.9), Hastings (0.85), Great Lakes (0.79), Port Stephens (0.67) and Bathurst (0.64) Why no regions like Dallas, Tampa-St Petersburg or Phoenix? All of NSW looks like San Francisco 28
NSW Planning System Legislation Planning and Environmental Assessment Act 1979 Environmental Planning and Assessment Regulation 2000 State Environmental Planning Policies (SEPPs) NSW-wide planning policies and procedures Local Environmental Plan (LEP) zones all land within an LGA; including what developments and land uses can occur and under what criteria LEPs need State Government approval 29
Some Unresolved Questions What criteria do Local Councils use to decide on the quantity of land to zone as residential in their LEPs? Answer seems to be that they use population growth projections What is the current stock of unused land that is currently zoned as residential in regional LGAs? Difficult to calculate this figure 30
Additional Slides 31
Elasticity Supply Elasticity and Area of LGA 1 0.8 y = -0.066x + 0.7638 R² = 0.124 0.6 0.4 0.2 0 0 2 4 6 8 10 12-0.2-0.4-0.6 Log of Area 32
Elasticity Estimates of Supply Elasticity by LGA 1 0.8 0.6 0.4 0.2 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100-0.2 33