国民経済計算における持ち家の帰属家賃推計について

Size: px

Start display at page:

Download "国民経済計算における持ち家の帰属家賃推計について"

ゆりかはまもり
7 years ago
Views:

1 ESRI Discussion Paper Series No.141 by May 2005 Economic and Social Research Institute Cabinet Office Tokyo, Japan

3 * * SNA

4 ( ) GDP 20 (1) (2) (3)

5 SNA ILO CPI GDP GDP

6 Estimating Imputed Rent for Owner-occupied Dwellings in the Japanese System of National Accounts * HaruhitoArai May 2005 * A preliminary version of this paper was presented to Sub-committee for Methodological Study in Advisory Committee for SNA on February 25, A summary version was also presented to Study Group on Housing Economy, organized by Housing Research & Advancement Foundation of Japan, on January 24, 2005, and to ESRI (Economic and Social Research Institute) International Conference on Next Steps for the Japanese System of National Accounts (Internal Session) on March 24, I thank participants of these meetings and Dr. Nobuo Akai, Associate Professor, School of Business Administration, University of Hyogo, for their valuable comments and suggestions. I also thank Messrs. Shuji Hasegawa, Kosuke Suzuki, and Toshitaka Nakano (National Expenditures Division, ESRI), Toshikazu Amari (National Product Division, ESRI), and Akira Takami (Consumer Statistics Division, Bureau of Statistics, Ministry of Internal Affairs and Communication) for their collaboration and comments. The views expressed in this paper are those of the author and do not necessarily represent those of ESRI or of the Japanese Government. Executive Research Fellow, Economic and Social Research Institute, Cabinet Office.

7 Abstract This paper analyses possible causes of large discrepancies among official data for imputed rent for owner-occupied dwellings in Japan, those including National Accounts, Regional Accounts, Input-output Table, and National Survey on Family Income Expenditure. By reviewing the methodologies used by each statistical agency in the light of academic researches on imputed rent and rent functions, this paper finds that imputed rent for owner-occupied dwellings is likely to be overestimated when average unit rent which reflects high unit rent for small dwellings in downtown is used to calculate imputed rent for large owner-occupied dwellings mainly located in the suburbs, and conversely, that imputed rent for owner-occupied dwellings is likely to be underestimated if rent function is used to calculate imputed rent for owner-occupied dwellings, and if the function faultily captures an apparent negative correlation between unit rent and floor space when location differences are not fully controlled for. This paper, using over 200 thousand realty rental advertisements in Metropolitan Tokyo, downloaded via Internet, then, examines the effectiveness of various estimation techniques used to control for location differences in estimating rent equations. Major findings include; i) stratification by area may not improve the performance of the set of area-wise rent functions; ii) use of area dummies in pooled estimate may not improve much the performance of the estimated rent function, either; and iii) alternative specifications can significantly improve the performance of rent functions, regardless of whether samples are stratified by area or not. To conclude, this paper proposes that i) National Accounts estimates of imputed rent for owner-occupied dwellings are calculated based on regional estimates, ii) in both National and Regional Accounts, imputed rent is calculated for each category of building structures, iii) National Survey of Family Income and Expenditure utilizes detailed information on housing location already available in Housing and Land Survey, or employs alternative specifications of rent function. Large discrepancies among official data for imputed rent for owner-occupied

8 dwellings in Japan are expected to narrow significantly if these proposals are adopted by statistical agencies. The final goal is that one single estimate of imputed rent for owner-occupied dwellings which utilize best available information provide the common benchmark for all other statistics.

9 15 2, GDP SNA 12 1

10 m 2 m 2 ( * m 2 m , , * 93SNA

11 59,168 44,560 7,145 6,904 ( ) 1,111, , , ,662 (m 2 ) () 23,432 13,138 57,592 58,285 14,986 1,036 56,601 57,310 8,446 12, ,334 1, ,112 10, [ ] 52,595 79, (1) (2) 3 (3) 3 (4) 4 * ln y ( i ) = a i + b ij x j + j c i ln S i y x j S m 2 a i,b ij,c j 1 m m 2 1,194 m 2 4 (1) * 11 VIII 3

12 (4) NEEDS-RADER

13 1989 NEEDS-RADER NEEDS-RADER PER PERPrice Earnings Ratio %

14 No m m (21.377)(2.01) u () (-6.131) (-4.907) R R^ = (11.409) *** (4.273) *** (11.609) *** u () (-5.197) *** (-7.556) *** R , adjustedr , F(4,285)88.00, Prob>F=0.00 6

15 ( ) (1 ) (10.633) *** (3.169) *** (11.225) *** exp( ) ( )0.7777( ) u (-2.891) *** (-8.312) *** (5.037) *** () R , adjustedr , F(5,286)69.58, Prob>F= i m 2 ii m 2 iii m 2 51 m 2 7

16 WLS i ii 51 m 2 iii 51 m ***(0.0067) ***(0.0077) 7.994***(0.0193) ***(0.0017) ***(0.0020) -0.15*** (0.002) ***(0.0009) ***(0.0009) ***(0.0016) ***(0.0015) ***(0.0015) ***(0.0025) 0.849***(0.0012) 0.646***(0.0017) 1.326***(0.0038) ***(0.0006) ***(0.0006) ***(0.0011) 0.572***(0.0798) (0.2208) 2.004***(0.0772) D (0.0126) * (0.0482) (0.0095) D ** (0.0088) (0.0231) (0.0068) D ***(0.0148) (0.0439) ***(0.0112) D-0.030* (0.0162) (0.0421) ***(0.0162) D-0.030***(0.0061) (0.0182) ***(0.0046) AdjR F D ****** No m 2 m 2 8

17 m 2 m (1)(5) 27 (1) 1 (2) m 2 m 2 m 2 m 2 * 9

23 N=196,590 N=102,136 N=94,454 Ln (rent) = 9.158507 + 0.

18 23 N=196,590 N=102,136 N=94,454 Ln (rent) = Ln (floor) Ln (rent) = Ln (floor) Ln (rent) = Ln (floor) 10

19 23 N=102,136 N=94,454 Ln (rent) = dum_ dum_ Ln (floor) 10 11

20 ) m 2 12

N=196,590 23 N=102,136 N=94,454 unit rent = 1119.085 + 8.615180 (floor) + 41205.

21 N=196, N=102,136 N=94,454 unit rent = (floor) (1/floor) unit rent = (floor) (1/floor) unit rent = (floor) (1/floor) 13

22 (16.1) 2.2 (14.8) 2.2 (13.5) 2.6 (15.1) 2.9 (16.1) 2.9 (15.5) 2.8 (14.4) 2.6 (12.6) 2.7 (12.6) 3.6 (16.1)

23 SNA SNA 50 SNA SNA SNA CD-ROM GDP SNA 15

24 SNA SNA

25 (1) (2) 3 (3)4 3 (1) (2) 3 (3) 3 (4) (2 ) (79,193) (57,310) m 2 (132.6) 29,160 38,420 63,798 87,858 80,774 m (597.2) m 2 (48.0) 23,718 34,019 41,842 58,871 64,670 m ,085.1 (,194.0)

26 (133.8) 12.7 (127.0) 14.8 (84.1) 18.1 (70.4) 24.7 (98.0) ( ) ( 1,943 ) 2,165 2,295 2,438 2, I % 18

28 (1) (2) (1) (2) GDP

29 21

30 93SNA VI IX IX 7 3 2ILO CPI ILO CPI CPI 93SNA CPI CPI 22

31 CPI security of tenure ability to modify the dwelling ILOCPI GDP (2002) (2003) (2002) (2003) (2003) (2001) (2002) GDP National Accounts of OECD Countries, Detailed Tables, Volume II, GDP IMF, World Economic Outlook, September % 15 23

32 5 GDP GDP GDP 24

33 15 GDP CD-ROM SNA 8 3 SNA SNA ILO2004, CONSUMER PRICE INDEX MANUAL: THEORY AND PRACTICE, Chapter 10 SOME SPECIAL CASES 15OECD2004, National Accounts of OECD Countries, Detailed Tables, Volume II, IMF, World Economic Outlook, September

34 No vol. 12, No.1, No No Autumn 26

untitled

untitled 2009 57 2 393 411 c 2009 1 1 1 2009 1 15 7 21 7 22 1 1 1 1 1 1 1 1. 1 1 1 2 3 4 12 2000 147 31 1 3,941 596 1 528 1 372 1 1 1.42 350 1197 1 13 1 394 57 2 2009 1 1 19 2002 2005 4.8 1968 5 93SNA 6 12 1 7,