The 25 references with contexts in paper Tara M. Sinclair (2012) “Characteristics and Implications of Chinese Macroeconomic Data Revisions” / RePEc:gwi:wpaper:2012-09

1
Ashiya, Masahiro. 2005. “Testing Rationality of Forecast Revision made by IMF and OECD.” Kobe University Discussion Paper No. 431 / 2005-02.
Total in-text references: 1
  1. In-text reference with the coordinate start=26436
    Prefix
    The inclusion of the constant, α, allows for the average revision to be nonzero. A statistically significant constant would suggest that the estimates are inefficient, but as discussed in
    Exact
    Ashiya (2005),
    Suffix
    this might occur in forecasts if the economy is frequently hit with unpredictable shocks with a common sign. Similarly for the estimates, if the data only available with a delay have a faster (slower) growth rate than the immediately available data, then we may expect a positive (negative) constant.

3
Bradsher, Keith. 2011. “China, Driver of World Economy, May be Slowing” The New York Times, September 23, 2011.
Total in-text references: 1
  1. In-text reference with the coordinate start=3821
    Prefix
    Although there has been substantial research on the accuracy of Chinese macroeconomic data generally, there has been little analysis of the revisions of the Chinese data 1 For example, see the New York Times article, “China, Driver of World Economy, May Be Slowing”
    Exact
    (Bradsher, 2011) and
    Suffix
    the 2011 speech by Justin Lin, Chief Economist and Senior Vice President of the World Bank, where he claims that “growth in China has been a driving force for the recovery from the global crisis since 2009,” and “in the decade beginning in 2000, China became the top contributor to the growth of global GDP.” 2 For a recent discussion of the literature on the quality of Chinese macro data, see Ap

5
Chow, G. 2006. “Are Chinese Official Statistics Reliable?” CESifo Economic Studies, Vol.52 2/2006, 396-414.
Total in-text references: 1
  1. In-text reference with the coordinate start=35633
    Prefix
    Like the U.S., it appears the Chinese official data can serve as “a reliable guide” to the level and growth pattern of GDP (Lequiller and Blades, 2007) and should be the “first port of call” (Scheibe, 2003). Alternative data resources have not proved to be more precise
    Exact
    (Holz 2006, Chow 2006).
    Suffix
    Furthermore, Klein and Ozmucur (2003) also found, based on principal components analysis, that the official real GDP estimates for China appear to be consistent with a set of indicators that captures the Chinese economy more broadly.

6
Croushore, D. 2011. "Frontiers of Real-Time Data Analysis," Journal of Economic Literature. 49(1): 72–100.
Total in-text references: 4
  1. In-text reference with the coordinate start=2067
    Prefix
    All remaining errors are the responsibility of the author. I. Introduction Traditional macroeconomic analysis assumes data revisions are small and random and thus have no effect on economic modeling, policy, or forecasting
    Exact
    (Croushore, 2011).
    Suffix
    Recent studies of U.S. macroeconomic data, however, suggest that revisions may be much more important than previously assumed (e.g. Aruoba, 2008; Croushore, 2008 and 2011; Dynan and Elmendorf, 2001; Kennedy, 1990; Sinclair and Stekler, 2012; Swanson and van Dijk, 2006).

  2. In-text reference with the coordinate start=6202
    Prefix
    Given the growing importance of the Chinese macroeconomy for global business, this paper fills a gap in the literature by providing a careful examination of China’s macroeconomic data revisions. The methodology follows the analysis of revisions previously applied to U.S. data (see
    Exact
    Croushore, 2011,
    Suffix
    for a recent survey). 3 Two exceptions are Wu (2007) and Holz (2008) who both examine the impact of the first economic census, conducted in 2004, on the Chinese GDP data.

  3. In-text reference with the coordinate start=29183
    Prefix
    The state of the economy likely was not in the information set at the time of the initial estimates, but it has become common for researchers to consider if the revisions of macroeconomic data series are different in recessions as compared to expansions (see
    Exact
    Croushore, 2011, and Sinclair and Stekler, 2012,
    Suffix
    for examples). Unlike the U.S., China does not have any negative growth rate observations in this sample that would match the typical description of a recession. China has, however, experienced “growth slowdowns.

  4. In-text reference with the coordinate start=34935
    Prefix
    Conclusions Chinese macroeconomic data generally fail Aruoba’s (2008) tests for well-behaved data revisions. The problems, however, are not at all unique for China, since even U.S. data fail most of Aruoba’s tests. Therefore, as argued by
    Exact
    Croushore (2011),
    Suffix
    it is not necessarily an indictment of the statistical authorities that the revisions are not well-behaved, but rather a call for further research to improve data gathering processes for all countries.

7
Croushore, D. 2008. "Revisions To PCE Inflation Measures: Implications for Monetary Policy" Philadelphia Fed working paper 08-8.
Total in-text references: 2
  1. In-text reference with the coordinate start=2250
    Prefix
    Introduction Traditional macroeconomic analysis assumes data revisions are small and random and thus have no effect on economic modeling, policy, or forecasting (Croushore, 2011). Recent studies of U.S. macroeconomic data, however, suggest that revisions may be much more important than previously assumed (e.g. Aruoba, 2008;
    Exact
    Croushore, 2008 and
    Suffix
    2011; Dynan and Elmendorf, 2001; Kennedy, 1990; Sinclair and Stekler, 2012; Swanson and van Dijk, 2006). The recent global dependence on the macroeconomic performance of China, 1 as well as concerns about the quality of the data being released by Chinese statistical agencies,2 suggest that a thorough study of the macroeconomic data revisions for China is an impo

  2. In-text reference with the coordinate start=5351
    Prefix
    In addition, it has been shown that for some U.S. macroeconomic data, large revisions are more likely in recessions than in expansions (Dynan and Elmendorf, 2001; Swanson and van Dijk, 2006). Furthermore, biases in data estimation have been identified for a number of U.S. macroeconomic series (Kennedy, 1990;
    Exact
    Croushore, 2008; Sinclair and Stekler, 2012).
    Suffix
    4 This research suggests that even the U.S. macroeconomic data experience substantial revision and bias, therefore the presence of revisions and biases in Chinese data would not be in itself evidence of lower data quality.

8
Croushore, D. and T. Stark. 2001. “A Real-Time Dataset for Macroeconomists.” Journal of Econometrics 105 (2001): 111-130. Curtis, C. C. and N. Mark. 2010. “Business Cycles, Consumption and Risk-Sharing: How Different Is China?” NBER Working Paper No. 16154. Dynan, K. E. and D. W. Elmendorf. 2001. "Do Provisional Estimates of Output Miss Economic Turning Points?" Federal Reserve Board Finance and Economics Discussion Series No. 2001-52.
Total in-text references: 2
  1. In-text reference with the coordinate start=10447
    Prefix
    and are subject to changes when more information from financial data and administrative records are available.” It was therefore necessary to construct a dataset that reflected the data that were available at different moments in time. The dataset for China was constructed in a similar manner to the “real-time dataset” for the U.S. introduced by
    Exact
    Croushore and Stark (2001).
    Suffix
    Following the terminology of Croushore and Stark (2001), “vintages” were created which capture the data from each statistical yearbook. For example, the 2006 vintage is the data printed in the 2006 statistical yearbook.

  2. In-text reference with the coordinate start=10513
    Prefix
    data and administrative records are available.” It was therefore necessary to construct a dataset that reflected the data that were available at different moments in time. The dataset for China was constructed in a similar manner to the “real-time dataset” for the U.S. introduced by Croushore and Stark (2001). Following the terminology of
    Exact
    Croushore and Stark (2001),
    Suffix
    “vintages” were created which capture the data from each statistical yearbook. For example, the 2006 vintage is the data printed in the 2006 statistical yearbook. Data revisions result in different vintages having different values for the same category and date.

10
Heston, A. 2001. “Treatment of China in PWT 6” http://pwt.econ.upenn.edu/Documentation/China.pdf
Total in-text references: 1
  1. In-text reference with the coordinate start=3328
    Prefix
    The Economist, 2008; Holz, 2003 and 2006; Huang, 2011; Keidel, 2001 and 2007; Maddison, 1998 and 2006; Orlik, 2012; Rawski, 1976 and 2001; Rawski and Xiao, 2001; Ren, 2002; Wu, 2000; Xu, 2002; Young, 2003). For example,
    Exact
    Heston (2001,
    Suffix
    pg. 3) claims that there are “winds of falsification” that surround the Chinese macroeconomic data, particularly in the late-1990s. Although there has been substantial research on the accuracy of Chinese macroeconomic data generally, there has been little analysis of the revisions of the Chinese data 1 For example, see the N

11
Holden, K., & Peel, D. A. 1990. “On testing for unbiasedness and efficiency of forecasts.” Manchester School, 48, 120–127.
Total in-text references: 4
  1. In-text reference with the coordinate start=17604
    Prefix
    statistics of the first revisions, i.e. the difference between the real growth rates released in the statistical yearbook two years after and those from one year after the year of the economic activity they refer to. For both GDP and GNP the means are statistically significantly positive at the 10% level. 11 This suggests that the initial estimates are biased
    Exact
    (Holden and Peel, 1990).
    Suffix
    Looking at the breakdown into the sectors it is clear where the bias is coming from – only the tertiary sector has a statistically significant mean, and it is again positive. In fact, the mean absolute revision for the tertiary sector is the same as the mean error.

  2. In-text reference with the coordinate start=29885
    Prefix
    We then apply a methodology similar to Sinclair et al (2010) and estimate the following regression: 12 revisiont = α + βslowdownt + et (2) where slowdown is an indicator variable that takes the value 1 in the slowdown years as listed above, and 0 otherwise. We will call this a modified
    Exact
    Holden and Peel (1990)
    Suffix
    test because it allows for two different biases depending upon whether or not the economy is in a slowdown. The joint test is that α = 0 and β = 0 so that there is no bias under the null.

  3. In-text reference with the coordinate start=40885
    Prefix
    0.12) 0.01 (0.03) 0.08 (0.14) 2.03*** (0.35) Coefficient on lag (Newey-West SE) 0.35 (0.22) -0.08 (0.16) 0.02 (0.02) 0.36* (0.18) 0.08 (0.17) p-value of F-test >0.01*** >0.01*** 0.50 0.14 >0.01*** Sample 1994-2008 1983-2008 1986-2008 1995-2008 1994-2008 *, **, *** Statistically significantly different from zero at the 10%, 5%, and 1% level respectively Table 12: Modified
    Exact
    Holden Peel (1990)
    Suffix
    with Slowdown Dummy for First Revisions GDP GNP PRIMARY SECONDARY TERTIARY Constant (Newey-West SE) 0.30*** (0.09) 0.30 (0.18) -0.11 (0.14) 0.11 (0.13) 0.75*** (0.14) Coefficient on dummy (Newey-West SE) -0.14 (0.17) -0.30 (0.37) 0.08 (0.15) -0.10 (0.21) -0.10 (0.20) p-value of F-test 0.02** 0.15 0.46 0.66 >0.01*** Sample 1993-2010 1982-2010 1985-2010 1994-2010 1993-

  4. In-text reference with the coordinate start=41355
    Prefix
    0.18) -0.11 (0.14) 0.11 (0.13) 0.75*** (0.14) Coefficient on dummy (Newey-West SE) -0.14 (0.17) -0.30 (0.37) 0.08 (0.15) -0.10 (0.21) -0.10 (0.20) p-value of F-test 0.02** 0.15 0.46 0.66 >0.01*** Sample 1993-2010 1982-2010 1985-2010 1994-2010 1993-2010 *, **, *** Statistically significantly different from zero at the 10%, 5%, and 1% level respectively Table 13: Modified
    Exact
    Holden Peel (1990)
    Suffix
    with Slowdown Dummy for Final Revisions GDP GNP PRIMARY SECONDARY TERTIARY Constant (Newey-West SE) 0.96*** (0.18) 0.68*** (0.20) -0.22 (0.25) 0.26 (0.21) 2.52*** (0.18) Coefficient on dummy (Newey-West SE) -0.56*** (0.16) -0.15 (0.22) 0.17 (0.26) -0.25 (0.23) -1.06** (0.38) p-value of F-test >0.01*** >0.01*** 0.25 0.41 >0.01*** Sample 1993-2008 1982-2008 1985-2008 1

12
Holz, C. 2003. “’Fast, Clear and Accurate:’ How Reliable Are Chinese Output and Economic Growth Statistics,” The China Quarterly, 173: 122-163.
Total in-text references: 1
  1. In-text reference with the coordinate start=3121
    Prefix
    However, the quality of data released by the National Bureau of Statistics of China (NBS), China’s statistical authority, has regularly been questioned by both the media and researchers (e.g. The Economist, 2008;
    Exact
    Holz, 2003 and
    Suffix
    2006; Huang, 2011; Keidel, 2001 and 2007; Maddison, 1998 and 2006; Orlik, 2012; Rawski, 1976 and 2001; Rawski and Xiao, 2001; Ren, 2002; Wu, 2000; Xu, 2002; Young, 2003). For example, Heston (2001, pg. 3) claims that there are “winds of falsification” that surround the Chinese macroeconomic data, particularly in the late-1990s.

14
Holz, C. 2006. "China’s Reform Period Economic Growth: How Reliable Are Angus Maddison’s Estimates? Response to Angus Maddison’s Reply." Review of Income and Wealth, 52(3): 471-5.
Total in-text references: 4
  1. In-text reference with the coordinate start=6314
    Prefix
    The methodology follows the analysis of revisions previously applied to U.S. data (see Croushore, 2011, for a recent survey). 3 Two exceptions are
    Exact
    Wu (2007) and Holz (2008)
    Suffix
    who both examine the impact of the first economic census, conducted in 2004, on the Chinese GDP data. Wu (2007) concludes that the 2006 benchmark revised GDP estimates based on the economic census are “not less questionable” than the data available before the census (pg. 2).

  2. In-text reference with the coordinate start=6648
    Prefix
    3 Two exceptions are Wu (2007) and Holz (2008) who both examine the impact of the first economic census, conducted in 2004, on the Chinese GDP data. Wu (2007) concludes that the 2006 benchmark revised GDP estimates based on the economic census are “not less questionable” than the data available before the census (pg. 2).
    Exact
    Holz (2008)
    Suffix
    concludes that the benchmark revision “implies that Chinese statistics have to be taken with a rock of salt” (Holz’s italics, pg. 163). 4 A longstanding concern about data revisions is their impact on forecasts and forecast evaluation, for example see Stekler (1967), Joutz and Stekler (1998), Croushore and Stark (2002). 2 Furthermore, a comparison between revisions in China with those i

  3. In-text reference with the coordinate start=9431
    Prefix
    the estimates of annual GDP printed in the official yearbooks are made separately from the quarterly estimates (Lequiller and Blades, 2007). 7 Further detail on the NBS definitions of the industries is available on their website: http://www.stats.gov.cn/english/classificationsmethods/definitions/t20020419_402787584.htm 3 approach
    Exact
    (Holz, 2008),
    Suffix
    unlike the U.S. that relies primarily on the expenditure approach to calculating GDP. The production approach is also considered to be the most reliable estimate for China (Lequiller and Blades, 2007).

  4. In-text reference with the coordinate start=20467
    Prefix
    For the tertiary sector, however, there were never any negative revisions as reported in Tables 5 and 6. This may be in part due to the shorter sample for the tertiary sector as compared, in particular, to GNP, but it is also consistent with the discussion in
    Exact
    Holz (2008)
    Suffix
    that the NBS had a sense that it was under reporting tertiary sector value added. 8 c. Comparison with U.S. Data To give some context for interpreting the size of the Chinese revisions, Tables 7 and 8 provide descriptive statistics for what is generally considered to be high-quality data, i.e.

15
Holz, C. 2008. “China’s 2004 Economic Census and 2006 Benchmark Revision of GDP Statistics: More Questions than Answers?” The China Quarterly, 193: 150-163.
Total in-text references: 1
  1. In-text reference with the coordinate start=3144
    Prefix
    However, the quality of data released by the National Bureau of Statistics of China (NBS), China’s statistical authority, has regularly been questioned by both the media and researchers (e.g. The Economist, 2008; Holz, 2003 and 2006;
    Exact
    Huang, 2011; Keidel, 2001 and
    Suffix
    2007; Maddison, 1998 and 2006; Orlik, 2012; Rawski, 1976 and 2001; Rawski and Xiao, 2001; Ren, 2002; Wu, 2000; Xu, 2002; Young, 2003). For example, Heston (2001, pg. 3) claims that there are “winds of falsification” that surround the Chinese macroeconomic data, particularly in the late-1990s.

16
Huang, Y., 2011. “Misinterpreting China’s Economy” Wall Street Journal, August 25, 2011.
Total in-text references: 3
  1. In-text reference with the coordinate start=4255
    Prefix
    Justin Lin, Chief Economist and Senior Vice President of the World Bank, where he claims that “growth in China has been a driving force for the recovery from the global crisis since 2009,” and “in the decade beginning in 2000, China became the top contributor to the growth of global GDP.” 2 For a recent discussion of the literature on the quality of Chinese macro data, see Appendix 1-1 from
    Exact
    Jia (2011)
    Suffix
    as well as the references discussed further below. 1 as released by the NBS. The small amount of evaluation of past macroeconomic data revisions for China has been used to argue that large data revisions suggest poor data quality for China. 3 In contrast with Chinese data, even early releases of U.

  2. In-text reference with the coordinate start=29474
    Prefix
    Unlike the U.S., China does not have any negative growth rate observations in this sample that would match the typical description of a recession. China has, however, experienced “growth slowdowns.” Following
    Exact
    Jia (2011),
    Suffix
    we use the following dates for the slowdowns: 1980, 1984, 1989, 1996-1998 and 2008. We then apply a methodology similar to Sinclair et al (2010) and estimate the following regression: 12 revisiont = α + βslowdownt + et (2) where slowdown is an indicator variable that takes the value 1 in the slowdown years as listed above, and 0 otherwise.

  3. In-text reference with the coordinate start=35929
    Prefix
    Furthermore, Klein and Ozmucur (2003) also found, based on principal components analysis, that the official real GDP estimates for China appear to be consistent with a set of indicators that captures the Chinese economy more broadly.
    Exact
    Jia (2011)
    Suffix
    does not find evidence of any measurement errors or irregularity in the Chinese real GDP data. Curtis and Mark (2010) also find that the official Chinese macro data appear consistent with economic 15 theory.

18
Joutz, Frederick & H. O. Stekler, 1998. "Data revisions and forecasting," Applied Economics, vol. 30(8), pages 1011-1016.
Total in-text references: 1
  1. In-text reference with the coordinate start=3144
    Prefix
    However, the quality of data released by the National Bureau of Statistics of China (NBS), China’s statistical authority, has regularly been questioned by both the media and researchers (e.g. The Economist, 2008; Holz, 2003 and 2006;
    Exact
    Huang, 2011; Keidel, 2001 and
    Suffix
    2007; Maddison, 1998 and 2006; Orlik, 2012; Rawski, 1976 and 2001; Rawski and Xiao, 2001; Ren, 2002; Wu, 2000; Xu, 2002; Young, 2003). For example, Heston (2001, pg. 3) claims that there are “winds of falsification” that surround the Chinese macroeconomic data, particularly in the late-1990s.

21
Kennedy, James E. 1990. "An Analysis of Revisions to the Industrial Production Index." Board of Governors of the Federal Reserve System, Economic Activity Section, Working Paper 109.
Total in-text references: 1
  1. In-text reference with the coordinate start=35677
    Prefix
    Like the U.S., it appears the Chinese official data can serve as “a reliable guide” to the level and growth pattern of GDP (Lequiller and Blades, 2007) and should be the “first port of call” (Scheibe, 2003). Alternative data resources have not proved to be more precise (Holz 2006, Chow 2006). Furthermore,
    Exact
    Klein and Ozmucur (2003)
    Suffix
    also found, based on principal components analysis, that the official real GDP estimates for China appear to be consistent with a set of indicators that captures the Chinese economy more broadly.

22
Klein, L. R. and S. Ozmucur. 2003. “The estimation of China’s Economic growth rate” Journal of Economic and Social Measurement, 28(4): 187-202.
Total in-text references: 5
  1. In-text reference with the coordinate start=9153
    Prefix
    is broadly the service sector. 7 China’s official approach is the production 5 The dataset is available at http://home.gwu.edu/~tsinc/china_rtd.xlsx 6 There are also quarterly data available for a portion of the sample, but the estimates of annual GDP printed in the official yearbooks are made separately from the quarterly estimates
    Exact
    (Lequiller and Blades, 2007).
    Suffix
    7 Further detail on the NBS definitions of the industries is available on their website: http://www.stats.gov.cn/english/classificationsmethods/definitions/t20020419_402787584.htm 3 approach (Holz, 2008), unlike the U.

  2. In-text reference with the coordinate start=9632
    Prefix
    of the industries is available on their website: http://www.stats.gov.cn/english/classificationsmethods/definitions/t20020419_402787584.htm 3 approach (Holz, 2008), unlike the U.S. that relies primarily on the expenditure approach to calculating GDP. The production approach is also considered to be the most reliable estimate for China
    Exact
    (Lequiller and Blades, 2007).
    Suffix
    For this reason this analysis will focus on the three main sectors rather than the sub-components of the expenditure approach for evaluation. a. Real-Time Dataset As regularly noted in the China Statistical Yearbook (for example in 2006): “Data on gross domestic product and related indicators of the most recent year published in the yearbook are not final and a

  3. In-text reference with the coordinate start=16265
    Prefix
    example, see McCully and Payson’s (2009) preview of the latest U.S. benchmark revisions released in July of 2009. 10 There was also a first tertiary industry census that resulted in benchmark revisions in 1995, but these do not affect much of the sample investigated here. 6 value added for the primary, secondary, and tertiary sectors. Consistent with the assumptions of the OECD
    Exact
    (Lequiller and Blades, 2007),
    Suffix
    the assumption in this paper is that the more heavily revised data are of higher quality than the earlier releases. It is the first release, however, that firms, households, and policymakers generally use in decision-making.

  4. In-text reference with the coordinate start=24520
    Prefix
    In the early part of the GNP sample, however, there is more variation in the direction of the revision. For the primary sector data we can see in Figure 3 that there has been little revision in the later part of the sample, which is consistent with the discussion in
    Exact
    Lequiller and Blades (2007).
    Suffix
    For the secondary sector data in Figure 4 we see that the latest available data is essentially the same as the second release data for 1994 through 2004. From 2004 through 2010 the latest available data for the secondary sector were generally revised upward from both the first and second releases.

  5. In-text reference with the coordinate start=35473
    Prefix
    It is also a reminder to firms and policymakers dependent on the early data to expect there to be substantial revisions in future releases. Like the U.S., it appears the Chinese official data can serve as “a reliable guide” to the level and growth pattern of GDP
    Exact
    (Lequiller and Blades, 2007) and
    Suffix
    should be the “first port of call” (Scheibe, 2003). Alternative data resources have not proved to be more precise (Holz 2006, Chow 2006). Furthermore, Klein and Ozmucur (2003) also found, based on principal components analysis, that the official real GDP estimates for China appear to be consistent with a set of indicators that captures the Chinese econo

24
Lin, Justin Yifu. 2011. “China and the Global Economy” Luncheon Address at the 2011 Asia Economic Policy Conference at the Federal Reserve Bank of San Francisco. Transcript available at: http://www.frbsf.org/economics/conferences/aepc/2011/volume/Lin.pdf
Total in-text references: 1
  1. In-text reference with the coordinate start=3185
    Prefix
    However, the quality of data released by the National Bureau of Statistics of China (NBS), China’s statistical authority, has regularly been questioned by both the media and researchers (e.g. The Economist, 2008; Holz, 2003 and 2006; Huang, 2011; Keidel, 2001 and 2007;
    Exact
    Maddison, 1998 and
    Suffix
    2006; Orlik, 2012; Rawski, 1976 and 2001; Rawski and Xiao, 2001; Ren, 2002; Wu, 2000; Xu, 2002; Young, 2003). For example, Heston (2001, pg. 3) claims that there are “winds of falsification” that surround the Chinese macroeconomic data, particularly in the late-1990s.

30
Orlik, T. 2012. “Debating Lies, Damned Lies, and Chinese Economic Statistics.” The Wall Street Journal Thursday, July 26, 2012: C12. Rawski T. G.1976. “On the reliability of Chinese economic data” Journal of Development Studies, 1743-9140, Volume 12, Issue 4, 1976, Pages 438 – 441. Rawski, T. G. 2001. "What is happening to China's GDP statistics?" China Economic Review, 12(4): 347-354.
Total in-text references: 1
  1. In-text reference with the coordinate start=3270
    Prefix
    However, the quality of data released by the National Bureau of Statistics of China (NBS), China’s statistical authority, has regularly been questioned by both the media and researchers (e.g. The Economist, 2008; Holz, 2003 and 2006; Huang, 2011; Keidel, 2001 and 2007; Maddison, 1998 and 2006; Orlik, 2012; Rawski, 1976 and 2001; Rawski and Xiao, 2001;
    Exact
    Ren, 2002; Wu, 2000; Xu, 2002; Young, 2003).
    Suffix
    For example, Heston (2001, pg. 3) claims that there are “winds of falsification” that surround the Chinese macroeconomic data, particularly in the late-1990s. Although there has been substantial research on the accuracy of Chinese macroeconomic data generally, there has been little analysis of the revisions of the Chinese data 1

31
Rawski T. G and W. Xiao, 2001 “RoundTable on Chinese Economic Statistics Introduction” China Economic Review, 12(2001) 298-302.
Total in-text references: 1
  1. In-text reference with the coordinate start=35553
    Prefix
    Like the U.S., it appears the Chinese official data can serve as “a reliable guide” to the level and growth pattern of GDP (Lequiller and Blades, 2007) and should be the “first port of call”
    Exact
    (Scheibe, 2003).
    Suffix
    Alternative data resources have not proved to be more precise (Holz 2006, Chow 2006). Furthermore, Klein and Ozmucur (2003) also found, based on principal components analysis, that the official real GDP estimates for China appear to be consistent with a set of indicators that captures the Chinese economy more broadly.

32
Ren, Rouen. 2002. “How Inflated Is China’s GDP Statistics?—Comment on Methodological Issues of Two Studies on China’s GDP Statistics.” Economics Quarterly 2 (1):37–52.
Total in-text references: 4
  1. In-text reference with the coordinate start=2328
    Prefix
    Recent studies of U.S. macroeconomic data, however, suggest that revisions may be much more important than previously assumed (e.g. Aruoba, 2008; Croushore, 2008 and 2011; Dynan and Elmendorf, 2001; Kennedy, 1990;
    Exact
    Sinclair and Stekler, 2012;
    Suffix
    Swanson and van Dijk, 2006). The recent global dependence on the macroeconomic performance of China, 1 as well as concerns about the quality of the data being released by Chinese statistical agencies,2 suggest that a thorough study of the macroeconomic data revisions for China is an important endeavor.

  2. In-text reference with the coordinate start=5351
    Prefix
    In addition, it has been shown that for some U.S. macroeconomic data, large revisions are more likely in recessions than in expansions (Dynan and Elmendorf, 2001; Swanson and van Dijk, 2006). Furthermore, biases in data estimation have been identified for a number of U.S. macroeconomic series (Kennedy, 1990;
    Exact
    Croushore, 2008; Sinclair and Stekler, 2012).
    Suffix
    4 This research suggests that even the U.S. macroeconomic data experience substantial revision and bias, therefore the presence of revisions and biases in Chinese data would not be in itself evidence of lower data quality.

  3. In-text reference with the coordinate start=29183
    Prefix
    The state of the economy likely was not in the information set at the time of the initial estimates, but it has become common for researchers to consider if the revisions of macroeconomic data series are different in recessions as compared to expansions (see
    Exact
    Croushore, 2011, and Sinclair and Stekler, 2012,
    Suffix
    for examples). Unlike the U.S., China does not have any negative growth rate observations in this sample that would match the typical description of a recession. China has, however, experienced “growth slowdowns.

  4. In-text reference with the coordinate start=34113
    Prefix
    Table 17 shows that the noise/signal ratio for the US final revisions is the same as what Aruoba (2008) reported in his Table 1 (page 325). The mean is statistically insignificant for the US real output in both cases, 14 but Aruoba finds substantial bias for other U.S. variables (as do others, for example,
    Exact
    Sinclair and Stekler, 2012).
    Suffix
    Finally, for the third property, we find predictability of the revisions from either a lag of the revision or the value of the first release for either the first or final revision for all five Chinese series as well as for U.

33
Scheibe, J. 2003. “Chinese Output Gap During the Reform Period 1978-2002” Department of Economics Discussion Paper No. 179, University of Oxford, December.
Total in-text references: 1
  1. In-text reference with the coordinate start=29610
    Prefix
    China has, however, experienced “growth slowdowns.” Following Jia (2011), we use the following dates for the slowdowns: 1980, 1984, 1989, 1996-1998 and 2008. We then apply a methodology similar to
    Exact
    Sinclair et al (2010) and
    Suffix
    estimate the following regression: 12 revisiont = α + βslowdownt + et (2) where slowdown is an indicator variable that takes the value 1 in the slowdown years as listed above, and 0 otherwise.

34
Sinclair, T. M. and H.O. Stekler, 2012. “Examining the quality of early GDP component estimates,” International Journal of Forecasting, forthcoming.
Total in-text references: 1
  1. In-text reference with the coordinate start=6913
    Prefix
    Holz (2008) concludes that the benchmark revision “implies that Chinese statistics have to be taken with a rock of salt” (Holz’s italics, pg. 163). 4 A longstanding concern about data revisions is their impact on forecasts and forecast evaluation, for example see
    Exact
    Stekler (1967),
    Suffix
    Joutz and Stekler (1998), Croushore and Stark (2002). 2 Furthermore, a comparison between revisions in China with those in the U.S. will give a relative measure of the performance of China’s statistical authority as compared to what is typically considered the “gold standard” in data production.

36
Stekler, H. O. 1967. “Data revisions and economic forecasting.” Journal of the American Statistical Association, 62, 470–483.
Total in-text references: 1
  1. In-text reference with the coordinate start=3270
    Prefix
    However, the quality of data released by the National Bureau of Statistics of China (NBS), China’s statistical authority, has regularly been questioned by both the media and researchers (e.g. The Economist, 2008; Holz, 2003 and 2006; Huang, 2011; Keidel, 2001 and 2007; Maddison, 1998 and 2006; Orlik, 2012; Rawski, 1976 and 2001; Rawski and Xiao, 2001;
    Exact
    Ren, 2002; Wu, 2000; Xu, 2002; Young, 2003).
    Suffix
    For example, Heston (2001, pg. 3) claims that there are “winds of falsification” that surround the Chinese macroeconomic data, particularly in the late-1990s. Although there has been substantial research on the accuracy of Chinese macroeconomic data generally, there has been little analysis of the revisions of the Chinese data 1

37
Swanson, N. R. and D. van Dijk. 2006. "Are Reporting Agencies Getting It Right? Data Rationality and Business Cycle Asymmetry. " Journal of Business and Economic Statistics, 24 (January 2006), pp. 24-42.
Total in-text references: 2
  1. In-text reference with the coordinate start=6314
    Prefix
    The methodology follows the analysis of revisions previously applied to U.S. data (see Croushore, 2011, for a recent survey). 3 Two exceptions are
    Exact
    Wu (2007) and Holz (2008)
    Suffix
    who both examine the impact of the first economic census, conducted in 2004, on the Chinese GDP data. Wu (2007) concludes that the 2006 benchmark revised GDP estimates based on the economic census are “not less questionable” than the data available before the census (pg. 2).

  2. In-text reference with the coordinate start=6465
    Prefix
    The methodology follows the analysis of revisions previously applied to U.S. data (see Croushore, 2011, for a recent survey). 3 Two exceptions are Wu (2007) and Holz (2008) who both examine the impact of the first economic census, conducted in 2004, on the Chinese GDP data.
    Exact
    Wu (2007)
    Suffix
    concludes that the 2006 benchmark revised GDP estimates based on the economic census are “not less questionable” than the data available before the census (pg. 2). Holz (2008) concludes that the benchmark revision “implies that Chinese statistics have to be taken with a rock of salt” (Holz’s italics, pg. 163). 4 A longstanding concern about data revisions is their impact on forecasts

38
Wu, H. X. 2000. “China's GDP level and growth performance: alternative estimates and the implications.” Review of Income and Wealth, 46(4).
Total in-text references: 1
  1. In-text reference with the coordinate start=3270
    Prefix
    However, the quality of data released by the National Bureau of Statistics of China (NBS), China’s statistical authority, has regularly been questioned by both the media and researchers (e.g. The Economist, 2008; Holz, 2003 and 2006; Huang, 2011; Keidel, 2001 and 2007; Maddison, 1998 and 2006; Orlik, 2012; Rawski, 1976 and 2001; Rawski and Xiao, 2001;
    Exact
    Ren, 2002; Wu, 2000; Xu, 2002; Young, 2003).
    Suffix
    For example, Heston (2001, pg. 3) claims that there are “winds of falsification” that surround the Chinese macroeconomic data, particularly in the late-1990s. Although there has been substantial research on the accuracy of Chinese macroeconomic data generally, there has been little analysis of the revisions of the Chinese data 1

39
Wu, H. X. 2007. “The Chinese GDP Growth Rate Puzzle: How Fast Has the Chinese Economy Grown?” Asian Economic Papers 6(1): 1-23.
Total in-text references: 1
  1. In-text reference with the coordinate start=3270
    Prefix
    However, the quality of data released by the National Bureau of Statistics of China (NBS), China’s statistical authority, has regularly been questioned by both the media and researchers (e.g. The Economist, 2008; Holz, 2003 and 2006; Huang, 2011; Keidel, 2001 and 2007; Maddison, 1998 and 2006; Orlik, 2012; Rawski, 1976 and 2001; Rawski and Xiao, 2001;
    Exact
    Ren, 2002; Wu, 2000; Xu, 2002; Young, 2003).
    Suffix
    For example, Heston (2001, pg. 3) claims that there are “winds of falsification” that surround the Chinese macroeconomic data, particularly in the late-1990s. Although there has been substantial research on the accuracy of Chinese macroeconomic data generally, there has been little analysis of the revisions of the Chinese data 1