Transcript PowerPoint ****

The dynamics of retail real estate
market in Hong Kong
Sun Zhuo Xiao · Chau Kwong Wing
Content
•
•
•
•
•
Background
Development of Hypothesis
The methodology
Data characteristics and sources
Empirical results
The importance of retail sector in Hong Kong
• By the end of 2013, retail trade industry offers 9.8% job positions and makes 5.0%
contribution to the GDP.
450000
6.0%
400000
5.0%
350000
300000
4.0%
250000
3.0%
200000
150000
2.0%
100000
1.0%
50000
0
0.0%
1999
2000
2001
2002
2003
2004
retail sales(HK$million)
2005
2006
2007
2008
GDP in wholesale and retail trade
2009
2010
2011
The importance of the retail property sector
In the aspect of factor of production, in year 2012, the real estate sector contributes
5.8% to the total GDP.
In the stock market, the market share of the real estate sector is even larger, they
occupy 14% of the total market value in September of 2013.(Hong Kong Exchange
– HKEx).
What might be the driven force behind it?
450.0
The retail property price index
400.0
350.0
300.0
250.0
200.0
150.0
100.0
50.0
0.0
1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Individual Visit Scheme
 First introduced in four Guangdong cities on 28 July 2003
 Under the Scheme, around 270 million Mainland residents in the 49 cities who have
permanent residence identity are eligible to apply for it.
 According to the statistics released by the Immigration Department, the ratio of mainland
travelers under the scheme accounts for 35% of the total mainland travelers in year
2004,and the ratio continues rising up to 67.1% in the first seven months of 2013.
 Till July of 2013, the number of travellers under the scheme reached 116.8 million since the
scheme was launched.
The performance of the tourism sector
Yearly tourist capacities for mainland visitors and
non-mainland visitors
350,000,000
31% 40% 54% 57% 54% 55% 55% 57% 61% 63% 67% 71%
Tourist spending from mainland China and other
areas of the world
300000
33% 40% 51% 65% 58% 56% 56% 58% 62% 69% 69% 71% 75%
250000
300,000,000
250,000,000
200000
200,000,000
150000
150,000,000
100000
100,000,000
50000
50,000,000
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
0
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Mainland visitors
Non-mainland visitors
The mainland of China
South&Southeast Asia
North Asia
The Americas
Europe,Africa&the Middle East
Taiwan
Australia,New Zealand&Souht Pacific
Macao
Development of Hypotheses
• With the increasing travellers from mainland China, the demand for retail business increases, therefore,
the demand for the retail space increases correspondingly. At the same time, the supply of retail property
space is sticky because the construction work takes time. In the framework of the demand and supply
mechanism, the retail property price will increase subsequently.
 Hypothesis1
IVS has a positive impact on the retail property price in Hong Kong.
Development of Hypotheses
• The expensive locations for retail property not only exhibit good locations, but also represent
convenient transportation, centralized luxury items and brand name shops, abundant subsidiary
facilities and well-known reputation. The expensive locations can simply attract more travellers than
the cheap locations. It is reasonable to propose that the impact of the IVS policy will be distinct in
expensive locations compared with the cheap locations.
• Hong Kong Island and Kowloon districts are classified into the expensive locations and New
Territories is classified into cheap location.
Hypothesis 2
IVS has a differential impact on the retail property in the expensive locations and the cheap
locations.
Development of Hypotheses
It is reasonable that the purchasing power of the travelers coming at the initial stage of the IVS is
higher than the subsequent visitors. In addition, the first beneficiaries of IVS from mainland China are
wealthier residents compared with the second and third-tier cities. Their affordability and the
probability of buying luxury items are much higher. The luxury items and brand name shops are
basically located in the expensive locations of the retail property area.
• Hypothesis 2a
IVS has a stronger impact on prices of street level shops in expensive locations compared with those in
cheaper locations.
Development of Hypotheses
The time table of IVS implementation
Time
City
Jul 28th, 2003
Dongguan, Foshan, Zhongshan, Jiangmen
Aug 20th, 2003
Guangzhou, Shenzhen, Zhuhai, Huizhou
Sep 1st, 2003
Shanghai, Beijing
Jan 1st, 2004
Shantou, Chaozhou, Meizhou, Zhaoqing, Qingyuan, Yunfu
May 1st, 2004
Shanwei, Maoming, Zhanjiang, Shaoguan, Jieyang, Heyuan, Yangjiang
Jul
Nanjing, Suzhou, Wuxi, Hangzhou, Ningbo, Taizhou, Fuzhou
1st, 2004
Mar 1st, 2005
Tianjin, Chongqing
Nov 1st, 2005
Chengdu, Jinan, Dalian, Shenyang
May 1st, 2006
Nanchang, Changsha, Nanning, Haikou, Guiyang, Kunming
Jan 1st, 2007
Shijiazhuang, Zhengzhou, Changchun, Hefei, Wuhan
Development of Hypotheses
• With the spread of the IVS policy, less travellers from first-tier cities come and more
travellers from second and third tier cities come who are generally less wealthier.
The attraction of expensive locations becomes weak and the impact of the policy
declines faster in expensive locations.
• Hypothesis 2b
The effect of IVS declines more slowly in cheaper locations than expensive locations.
The methodology to conduct the study
• Ordinary least squares regression analysis in the preliminary process
OLS is an approach for estimating the unknown coefficients in a linear
regression model. Assume there is a single linear relationship among the
variables 𝑥1𝑡, 𝑥2𝑡 , … 𝑥𝑘𝑡 in the form of
• 𝑥1𝑡 =a+
𝑘
𝑗=2 𝑏1𝑗 𝑥𝑗𝑡
+ 𝑢𝑡
Data characteristics and sources
Variable
Description
Definition
Data Source
P
Price
Rating and valuation department
RRS
Real Retail Sales
Price for retail property in HK and three districts of
Hong Kong
Retail sales value in Hong Kong divided by CPI
SPA
Newly Completely Space
IVS
IVS*@trend
Individual Visit Scheme
Dummy variable
EXC
Effective Exchange rate Index
HIBOR3M
IS
Hong Kong Monthly digest of Statistics by Census and
Statistics Department
Completions based on the issue of an occupation Hong Kong Property Review by Rating and Valuation
permit
Department
Launched in July of 2003
Government Policy
Equal to 0 before 2003, and equal to 1 thereafter
The time trend of IVS
The effective exchange rate index (EERI) for the Hong Kong Monthly digest of Statistics
HKD is an index which measures movements in the
weighted average of the exchange rate of the HKD
against the currencies of major trading partners of
Hong Kong.
Hong Kong Interbank Offered As a proxy for interest rate
Hong Kong Monetary Authority
Rate 3 month
Income to space ratio
Income to space ratio is calculated as GDP divided by Hong Kong GDP : Hong Kong yearly digest of Statistics
the retail stock at the year end.
Stock completed at the yearend: Hong Kong Property
Review
V
Vacancy rate
Vacancy rate at year end
ST
Stock
Stock at year end
Hong Kong Property Review by Rating and Valuation
Department
Rating and Valuation deparetment
Empirical model
• ∆ ln P = λ0 + λ1 ∆ ln RRS(−1) + λ2 ∆ ln SPA(3) + λ3 IVS + λ4 IVS ∗ Trend + λ5 EXC +
λ6 HIBOR3M + λ7 IS + λ8 V + MA 1 + ε
(1)
H1
• ∆ ln HKIP = α0 + α1 ∆ ln RRS + α2 ∆ ln HKISPA(10) + α3 IVS +
α4 IVS ∗ Trend + α5 EXC + α6HIBOR3M + MA 1 + AR 1 +
ε
H2
(2)
• ∆ ln KLP = β0 + β1 ∆ ln RRS + β2 ∆ ln KLSPA(10) + β3 IVS + β4 IVS ∗ Trend + β5 EXC +
β6 HIBOR3M + AR 1 + MA(1) + ε
(3)
• ∆ ln NTP = δ0 + δ1 ∆ ln RRS + δ2 ∆ ln NTSPA(10) + δ3 IVS + δ4 IVS ∗ Trend + δ5 EXC +
δ6 HIBOR3M + AR 1 + MA 1 + ε
(4)
Empirical model
• The second step is to differentiate the price equation of the Hong Kong Island with the New
Territories and the price equation of Kowloon with New Territories.
• The two equations are achieved as
• ∆ ln HKIP − ∆ ln NTP = γ0 + γ1 ∆ ln RRS +
γ2 ∆ ln HKISPA 10 − ∆ ln NTSPA 10
γ6 HIBOR3M + AR 1 + MA 1 + ε
+ γ3 IVS + γ4 IVS ∗ Trend + γ5 EXC +
(5)
• ∆ ln KLP − ∆ ln NTP = ϕ0 + ϕ1 ∆ ln RRS + ϕ2 (∆ ln KLSPA 10 −
∆ ln NTSPA 10 ) + ϕ3 IVS + ϕ4 IVS ∗ Trend + ϕ5 EXC + ϕ6 HIBOR3M + AR 1 +
MA(1) + ε
(6)
Empirical result for Hypothesis 1
Variable
Intercept
∆lnRRS(-1)
∆lnSPA(3)
IVS
Exchange
IVS*@TREND
IS
V
∆lnHibor3M
MA(1)
Adjusted R-squared
Coefficient
0.945386
0.783715
-0.098693
2.664560
-0.011434
-0.108621
0.045353
-2.433189
-0.011186
-0.999791
0.8468
Prob.
0.0000
0.0308
0.0491
0.0000
0.0147
0.0000
0.0456
0.0708
0.0672
0.0126
Empirical result for Hypothesis 2
Hong Kong Island
Intercept
Kowloon
0.313590***
Intercept
(0.0000)
∆lnRRS
0.662637**
-0.087118***
∆lnRRS
0.352727***
∆lnKLSPA
-0.004780***
IVS
-0.002845***
IVS*@trend
-0.106405***
Exchange
0.308840***
∆lnHibor3M
-0.978014***
(0.0000)
0.530759***
-0.007462***
-0.002757***
-0.129705***
AR(1)
0.475426***
IVS
-0.999946***
(0.0000)
0.002263
(0.9858)
IVS*@trend
-0.000174
(0.9196)
Exchange
-0.001259*
(0.0529)
∆lnHibor3M
-0.002496
(0.9488)
AR(1)
(0.0000)
MA(1)
-0.007407
(0.8414)
(0.0009)
(0.0063)
MA(1)
∆lnNTSPA
(0.0051)
(0.0006)
AR(1)
-0.048317
0.962337***
(0.0195)
(0.0000)
(0.0000)
∆lnHibor3M
∆lnRRS
(0.0000)
(0.0010)
Exchange
0.800022***
0.148141***
(0.0372)
(0.1138)
(0.0008)
IVS*@trend
Intercept
(0.0310)
(0.0047)
IVS
0.304739***
(0.0049)
(0.0463)
∆lnHKISPA
New Territories
0.401008***
(0.0001)
MA(1)
-0.999925***
(0.0000)
Empirical results for Hypotheses 2a and 2b
Variable
Coefficient
Prob.
Variable
Coefficient
Prob.
C
0.059429
0.1715
C
0.117761
0.0429
DLOG(RRS)
0.274711
0.3756
DLOG(RRS)
-0.002603
0.9944
0.3599
DLOG(KLSPA(10
0.002606
0.8890
DLOG(HKISPA(10 -0.015488
))-
))-
DLOG(NTSPA(10)
DLOG(NTSPA(10
)
))
IVS
0.148877*
0.0789
IVS
0.354036***
0.0005
IVS*@TREND
-0.001996*
0.0832
IVS*@TREND
-0.004981***
0.0002
EXCHANGE
-0.000560
0.1645
EXCHANGE
-0.001090
0.0421
DLOG(HIBOR3M -0.049470
0.0482
DLOG(HIBOR3
-0.080765
0.0128
M)
)
AR(1)
0.028711
0.8121
AR(1)
0.118123
0.3467
MA(1)
-0.970422
0.0000
MA(1)
-0.999840
0.0000
Conclusions
• IVS has a positive impact on the retail property price in Hong Kong
Conclusions
• IVS has a differential impact on retail property price in the expensive
locations and cheap locations
• The impact is stronger in expensive locations
• The impact declines slowly in cheap locations
Thank you
&
Questions?