Housing, the NZ Business Cycle and Macro Im balances Peter Jarrett - - PowerPoint PPT Presentation

Housing, the NZ Business Cycle and Macro Im balances

Peter Jarrett OECD Economics Department OECD Economics Department 23 June 2011

Outline

F t f th t t NZ h i b

• 1. Features of the most recent NZ housing boom
• 2. How NZ housing cycles compare with those in
• ther OECD countries
• ther OECD countries
• 3. Role of housing in NZ macroeconomic cycles

and imbalances and imbalances

Housing investment and GDP over the longer term

Billions of NZD Billions of NZD

7000 8000

• A. Real housing investment

130 140 150

• B. Real gross domestic product

5000 6000 100 110 120 3000 4000 70 80 90 1970 1975 1980 1985 1990 1995 2000 2005 2010 2000 1970 1975 1980 1985 1990 1995 2000 2005 2010 50 60

F t f th t t NZ h i

• 1. Features of the m ost recent NZ housing

boom 2 How NZ housing cycles compare with those in

• 2. How NZ housing cycles compare with those in
• ther OECD countries

3 Role of housing in NZ macroeconomic cycles

• 3. Role of housing in NZ macroeconomic cycles

and imbalances

From 2000-2007, house prices increased in NZ more than in most other OECD countries

• om

240

New Zealand Australia

240

Canada Netherlands

Housing price indexes across countries

Index 2001=100

housing bo

180 200 220

Australia France Ireland United States

180 200 220

Netherlands Spain United Kingdom Belgium

he recent h

140 160 180 140 160 180

atures of th

100 120 100 120

Fea

2002 2004 2006 2008 2010 80 2002 2004 2006 2008 2010 80

… as did household debt

• om

Household debt

As a percentage of gross disposable income

housing bo

As a percentage of gross disposable income

he recent h atures of th Fea

Strong net migration and easy credit conditions were major drivers

• om

% of population

25

%

• A. Housing prices and net migration

180

NZD billions

• B. Lending by financial institutions

housing bo

0.8 1.0 1.2 15 20

g p g

House price growth¹ (right scale) Net migration (left scale) 120 140 160

g y

Housing Business Agriculture

he recent h

0.2 0.4 0.6 5 10 60 80 100

atures of th

1990 1995 2000 2005 2010

• 0.2

0.0

• 5

1998 2000 2002 2004 2006 2008 2010 20 40

1. House prices measured by the Quotable Value house price index, house only; % change from previous year. Source: ANZ; Reserve Bank of New Zealand; Statistics New Zealand; and OECD Economic Outlook 89 database.

Fea

; ; ; 9

NZ has experienced one of the OECD’s highest population growth rates

• om

Percentage population change 20 0 0 -20 0 9

10 15 10 15 Natural change Net migration share

housing bo

5 10 5 10

he recent h atures of th

• 5
• 5

POL DEU JPN BEL GRC FIN ITA GBR DNK KOR NLD AUT SWE FRA NOR CHE USA IRL CAN MEX NZL ESP ISL AUS

Fea

NZ migrants tend to be highly skilled

Share of m igrants that are highl skilled and em plo ed

30 30

Share of m igrants that are highly skilled and em ployed

Percent

• om

20 25 20 25

housing bo

10 15 10 15

he recent h

5 5

atures of th

TUR ITA PRT POL SVK CZE HUN MEX AUT GRC LUX ESP FRA DEU CHE GBR OECD NLD SWE AUS DNK BEL FIN IRL JPN NOR NZL USA CAN

Fea

… yet owner-occupancy rates continued their long-term decline over the boom period

NZ Owner-occupancy rates

• om

..suggesting reduced affordability and an increasing role for rental property investments

0 72 0.74 0.76

housing bo

0.68 0.7 0.72

he recent h

0.64 0.66 0.68

atures of th

0.62

Mar 91 Mar 92 Mar 93 Mar 94 Mar 95 Mar 96 Mar 97 Mar 98 Mar 99 Mar 00 Mar 01 Mar 02 Mar 03 Mar 04 Mar 05 Mar 06 Mar 07 Mar 08 Mar 09 Mar 10 Mar 11*

Fea

31 31 31 31 31 31 31 31 M 31 31 M 31 31 M 31 M 31 M 31 M 31 M 31 31 M 31 M 31 31 M

Source: Statistics NZ

Housing supply lagged

… .Number of dwellings per household began to fall in 2002; average house size is still trending up

• om

housing bo he recent h atures of th Fea

The response of real residential investment was short-lived and eventually fell well below long-run equilibrium levels b d OECD ti t t d i … based on OECD estimates presented in:  Caldera Sanchez and Johansson (2011): panel SUR estimations, ECM

• om

housing bo he recent h atures of th Fea

Housing is tax-preferred

• om

housing bo he recent h atures of th Fea

F t f th t t NZ h i b

• 1. Features of the most recent NZ housing boom
• 2. How NZ housing cycles com pare with the

international experience international experience

• 3. Role of housing in NZ macroeconomic cycles

and imbalances and imbalances

Does NZ face greater restrictions on housing supply?

Relative to 21 OECD countries:

• Residential investment responds to house prices at an average pace
• But population changes have a much larger impact on long-run house prices in NZ than

elsewhere

2.5

Estim ates of LR Housing Supply Response to Price (Caldera Sanchez and Johansson, 2011)

30

LR House Price response to Population (Caldera Sanchez and Johansson, 2011)

us OECD

1 1.5 2 15 20 25

sing versu

0.5 CHE NLD AUT ITA BEL FRA ISR UK DEU POL ESP NOR AUS IRL NZL FIN JPN CAN DNK SWE USA 5 10 D… TA N… ESP USA UT FIN SR PN OL AN D… RA S… RL HE US LD EL UK NZL

NZ hous

IT E U AU F I JP PO CA FR I CH AU NL B U N

Does NZ face greater restrictions on housing supply?

Supply responsiveness ¹

S

Supply responsiveness ¹

S

1.6 1.8 2.0

USA

Correlation coefficient = -0.56

1.6 1.8 2.0

USA

Correlation coefficient = -0.45

us OECD

1.2 1.4

CAN DNK SWE

1.2 1.4

CAN DNK SWE

sing versu

0.6 0.8 1.0

AUS IRL

New Zealand

FIN JPN

0.6 0.8 1.0

AUS IRL

New Zealand

FIN JPN

NZ hous

0.2 0.4

CHE NLD AUT ITA BEL FRA GBR DEU POL ESP NOR AUS ISR

0.2 0.4

CHE NLD AUT ITA BEL FRA GBR DEU POL ESP NOR AUS ISR

50 100 150 200 0.0

Number of days to obtain a building permit ²

200 400 0.0

Population density³

But regulation may be discouraging buy-to-let investment

How do NZ housing cycles compare with OECD experience?

• Examine all major upturns and downturns in 18 OECD

countries since 1970

– André (2010): defined by real house price change > 15%

• New Zealand episodes identified:

us OECD

• New Zealand episodes identified:

Upturns Downturns Q Q % Q 8 Q 8% sing versu 1970Q1-1974Q3: +73% 1974Q3-1980Q4: -38% 1980Q4-1984Q2: +33% 1986Q4-1989Q1: +15% Q Q % NZ hous 1992Q1-1997Q3: +40% 2000Q4-2007Q3: +98%

Real house price appreciation in upturns has tended to be positively associated with population growth across countries

House price changes vs. population changes

250

% change in real house prices over upturns

• 10

% change in real house prices over downturns

200 250

• 20

10

us OECD

100 150

NZL

• 40
• 30

NZL

sing versu

50 100

NZL NZL NZL NZL

• 50

40

NZL

NZ hous

5 10 15 20

% population change over upturns

NZL

5 10

• 60

% population change over downturns

… and negatively with LT interest-rate changes

250

% change in real house prices over upturns

10

% change in real house prices over downturns

House-price changes vs. interest-rate changes

200 250

• 20
• 10

us OECD

100 150

NZL

• 40
• 30

NZL

sing versu

50 100

NZL NZL NZL NZL

• 50
• 40

NZL

NZ hous

• 10
• 5

5

% point change in long-term interest rates over upturns

NZL

• 5

5 10

• 60

% point change in long-term interest rates over downturns

During upturns, the share of residential investment in GDP has tended to increase more in NZ than elsewhere

% change in real house prices over upturns % change in real house prices over downturns

House price changes vs. change in residential investm ent (% GDP)

200 250

% change in real house prices over upturns

• 20
• 10

% change in real house prices over downturns

us OECD

150

• 30

sing versu

50 100

NZL NZL NZL NZL

• 50
• 40

NZL

NZ hous

• 5

5

% point change in share of residential invest

NZL NZL

• 8
• 6
• 4
• 2
• 60

% point change in share of residential invest

… but its dwelling stock tended to grow less than average for each % increase in population

90

% change in dwelling stock over upturns

90

% change in dwelling stock over upturns

Change in dwelling stock during upturns

70 80 90 70 80 90

us OECD

40 50 60 40 50 60

sing versu

20 30

NZL

10 20 30

NZL NZL

NZ hous

100 200 10

% change in real house prices over upturns

NZL NZL

5 10 15 20

% change in population over upturns

NZL

F t f th t t NZ h i b

• 1. Features of the most recent NZ housing boom
• 2. How NZ housing cycles compare with those in
• ther OECD countries
• ther OECD countries
• 3. Role of housing in NZ m acroeconom ic

cycles and im balances cycles and im balances

Most of the deterioration in NZ’s external imbalance since the early 1990s reflects private-sector outcomes …

10 10

alances

Sectoral contributions to the current account

Per cent of GDP

5 10 5 10

• mic Imba
• 5
• 5

acroecono

• 10
• 10

Government Private Current account balance

ing and M

1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010

• 15
• 15

Housi

Indeed, it is on the saving side that the distinction lies

alances Net saving rate

Average 2004-10, per cent of GDP

• mic Imba

15 20 25 15 20 25

♦ ♦ ♦ ♦ ♦ ♦

National Private Public

acroecono

5 10 5 10

♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦

ing and M

• 10
• 5
• 10
• 5

♦ ♦ ♦

GRC ISL PRT USA SVK HUN NZL ITA GBR JPN POL ESP FRA CZE AUS DNK BEL DEU ISR FIN SVN CAN EST AUT IRL NLD SWE CHE MEX KOR NOR

Housi

… private business investment has also underperformed

Investm ent rates

alances

Investm ent rates

Per cent of GDP

30

• A. Total investment

New Zealand Australia

10

• B. Public investment

OECD median¹

• mic Imba

20 25 6 8

acroecono

22 8 1980 1985 1990 1995 2000 2005 2010 15 20 1980 1985 1990 1995 2000 2005 2010 2 4

ing and M

16 18 20 22

• D. Business fixed investment

6 7 8

• C. Residential investment

Housi

8 10 12 14 3 4 5 1980 1985 1990 1995 2000 2005 2010 8 1980 1985 1990 1995 2000 2005 2010 3

Household saving has been negative since the early 1990s

8 8

Business Household

alances

Private-sector saving rates

As a percentage of national disposable income, 3-year average

2 4 6 2 4 6

Household

• mic Imba
• 2

2

• 2

2

acroecono

1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010

• 6
• 4
• 6
• 4

ing and M Housi

Source: Savings Working Group (2011)

… how much of household saving behaviour can be explained by the housing boom?

alances House price changes vs. change in household saving rate

% i h i HH i i % i h i HH i i d

• mic Imba

5 10

% point change in HH saving ratio over upturns

10 15

% point change in HH saving ratio over downturns

acroecono

NZL NZL

5

ing and M

• 10
• 5

NZL

• 5

Housi

50 100 150 200

• 15

% change in real housing price

• 50
• 40
• 30
• 20
• 10
• 10

% change in real housing price

Panel estimations of the household saving rate

Based on the ARDL framework of Roehn (2010)

Long run

Estimates for All Estimates for NZ Estimates for NZ

Long run

Estimates for All Estimates for NZ Estimates for NZ

Structural Fiscal balance

• 0.34*
• 0.56*
• 0.59***

(0.21) (0.3) (0.18)

Log of real household disposable income per capita

• 34.87
• 14.93***
• 14.70***

(22.34) (4.83) (4.02)

alances

(22.34) (4.83) (4.02)

Inflation

95.20* 38.51** 40.00***

(48.99) (15.60) (12.64)

Old age dependency ratio

• 3.54
• 0.26

(3.70) (1.51)

Real short-term interest rate

0 09

• 0 03
• mic Imba

Real short term interest rate

0.09 0.03

(0.08) (0.11)

Short run Error correction term

• 0.20***
• 0.20***
• 0.20***

(0.03) (0.03) (0.03)

Lagged ΔHH saving rate

0 14* 0 85*** 0 85***

acroecono

Lagged ΔHH saving rate

0.14* 0.85*** 0.85***

(0.08) (0.06) (0.06)

ΔStructural fiscal balance

• 0.27***
• 0.34***
• 0.25**

(0.07) (0.12) (0.10)

ΔLog of real GDP per capita

• 15.49***
• 5.97

ing and M

(4.09) (3.95)

ΔLog of real house price

• 3.62*
• 7.46*
• 7.98***

(2.05) (3.87) (2.79)

ΔLog of real stock prices

• 0.26
• 0.02

(0.21) (0.62)

Housi

ΔTerms of trade

0.01 0.02

(0.01) (0.02)

constant

20.20** 29.32*** 28.09***

(10.22) (7.39) (7.27)

These results suggest NZ’s equilibrium household saving rate trended downwards until recently, even without accounting for house prices

alances

4 8

• mic Imba

nt

acroecono

• 4

Perce

ing and M

• 8

Housi

• 12

86 88 90 92 94 96 98 00 02 04 06 08

Household saving rate g Estimated long-run equilibrium household saving rate

Panel estimations on household saving rate: results for NZ

alances

3 8 Percent Percentage points

• mic Imba

1 2 3 4 8

acroecono

• 2
• 1
• 8
• 4

ing and M

• 4
• 3

86 88 90 92 94 96 98 00 02 04 06 08

Change in household saving rate

• 12

86 88 90 92 94 96 98 00 02 04 06 08

Household saving rate

Housi

In-sample predicted change Dynamic in-sample predicted value

Simulation: How NZ’s household saving rate would have evolved if real house prices stayed flat at 2000 values

8 4 8

alances

cent

• mic Imba
• 4

Perc

acroecono

• 12
• 8

ing and M

12 86 88 90 92 94 96 98 00 02 04 06 08 Household saving rate Dynamic in-sample predicted value

Housi

y p p Simulation: Real house prices flat since 2000

Projecting NZ’s household saving rate: a V-shaped pattern

alances ..assuming all input variables follow the path of OECD Economic Outlook projections and real house prices stay flat, the household saving rate would improve in 2012 (though beware of earthquake effects)

• mic Imba

110 120

NZ Real Housing Price Index

2

NZ Household Saving Rate

acroecono

80 90 100

• 6
• 4
• 2

ing and M

50 60 70

• 12
• 10
• 8

Housi

00 01 02 03 04 05 06 07 08 09 10 11 12 Source: QVNZ up to 2010Q4 00 01 02 03 04 05 06 07 08 09 10 11 12

A summary

• The 2000-07 boom was led by population growth and easy credit and led to declining
• wner-occupancy. The dwelling stock per household has fallen steadily, as supply has

d d l kl t th i ti l ith t bl k i 8 responded only weakly to the price stimulus, with notable weakness since 2008.

• House prices respond unusually heavily to population more generally; only once

prices have risen does supply react. What precisely are the main barriers to supply

• ther than possibly landlord-tenant regulation and zoning restrictions in some areas

is not entirely clear.

• NZ ‘s numerous housing price upturns have been characterised by unusually small

changes in long-term interest rates, moderate volatility in the share of residential investment in GDP but small increases in the dwelling stock. g

• NZ’s chronic external imbalance results from weak private and especially household

saving and, to a lesser extent, from high public investment.

• Low household saving and its focus on property components is exacerbated by the

tax treatment of saving But low saving can be more fundamentally attributed to tax treatment of saving. But low saving can be more fundamentally attributed to trend declines in inflation, improvements in the public finances and gains in real per capita incomes and, indeed, housing price appreciation. The household saving rate should imminently enjoy a fairly sharp rebound, assuming flat house prices.