Challenges of working with the Chinese NBS firm-level data

Brandt, Loren; Biesebroeck, Johannes Van; Zhang, Yifan

doi:10.1016/j.chieco.2014.04.008

Cited by 220 publications

(96 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There is some evidence that the share of intermediates may be higher in China than the US, see, e.g.,Table 1inBrandt et al (2014). We re-computed the bound withζ = 0.25 and obtained very similar results.…”

mentioning

confidence: 62%

“…The original data come as a repeated cross-section. A panel is constructed following almost directly the method outlined in Brandt et al (2014), which also contains an excellent overview of the data for the interested reader. The Chinese data have been used multiple times and are by now familiar in the misallocation literaturefor example, Hsieh and Klenow (2009) -although our use of the panel dimension is rather new.…”

Section: Datamentioning

confidence: 99%

See 1 more Smart Citation

The Sources of Capital Misallocation

David

Venkateswaran

2019

American Economic Review

163

View full text Add to dashboard Cite

We develop a methodology to disentangle sources of capital “mis-allocation,” i.e., dispersion in value-added/capital. It measures the contributions of technological/informational frictions and a rich class of firm-specific factors. An application to Chinese manufacturing firms reveals that adjustment costs and uncertainty, while significant, explain only a modest fraction of the dispersion, which stems largely from other factors: a component correlated with productivity and a fixed effect. Adjustment costs are more salient for large US firms, though other factors still account for the bulk of the dispersion. Technological/markup heterogeneity explains a limited fraction in China, but a potentially large share in the United States. (JEL D22, D24, D25, E22, G31, L60, O11, O14, O47, P31)

show abstract

mentioning

confidence: 62%

Section: Datamentioning

confidence: 99%

The Sources of Capital Misallocation

David

Venkateswaran

2019

American Economic Review

163

View full text Add to dashboard Cite

show abstract

“…Between NECs, annual data are collected only through sampling of smaller firms and reporting from large businesses fulfilling certain changeable criteria, by both provincial statistical agencies and the NBS. This creates sampling biases and inconsistencies 6,9,20 . Moderate revisions for the years between one NEC and the next are thus expected, but coal consumption has typically been revised by 5-10%-even more in the latest NEC-and almost exclusively upwards.…”

Section: Figure 1 | Comparison Of Growth Rates For DI Erent Measures mentioning

confidence: 98%

Uncertainties around reductions in China’s coal use and CO2 emissions

2016

View full text Add to dashboard Cite

, implying high uncertainty. For example, BP 15 used different Chinese data to estimate a 0.9% increase in 2014 CO 2 emissions 7,8 . Here, we analyze 16 these preliminary announcements, with an approach that can be used to assess robustness of similar 17 future announcements. We show that the preliminary 2.9% reduction in coal consumption is 18 . Because it refers to coal consumption measured in mass units and often deviates from 9 growth measured in energy units, it can be misleading for estimating CO 2 emissions. It is often revised 10 ( Figure 1), but usually not until about 18 months later, with the publication of the following year's China 11 Energy Statistical Yearbook. A means of assessing the robustness of this growth rate is therefore 12 desirable, which we provide later in this paper. 13Further energy consumption data later published by NBS (and used by BP) show an insignificant increase 14 (0.06%) in total coal-derived energy use in 2014 7,8,10 , which is measured in energy units rather than mass 15 units, and includes energy from imports/exports and stock changes of coke and other products derived 16 from coal (see Methods online). Such moderate differences between growth rates of coal consumption 17 in mass units and growth rates of total coal-derived energy use are not uncommon in Chinese data, and 18 they have increased in recent years ( Figure 1). Coal-derived energy use is the most appropriate quantity 19 to use for estimating CO 2 emissions, since it explicitly includes all coal-derived products, and because 20 carbon content correlates more closely with energy content than with the mass of the coal 11 . Using 21 NBS's energy consumption data for coal, oil and natural gas 10 , combined with mean emission factors and 22 oxidation rates from a recent study (Liu et al. 2015) 12 , we estimate that Chinese fossil CO 2 emissions 23 3 grew by 0.8% in 2014 (see Methods). Global estimates compatible with this data show global energy-1 related emissions likely went up by ~0.5%, much less than the average over the last decade 7,13 . 2Interpretation of Chinese coal statistics is further complicated by revisions after the recently conducted 3 third National Economic Census (NEC), on which data in the 2014 Statistical Communiqué and 4 subsequently published data are based. The third NEC collected comprehensive data on economic 5 activity and energy use from (in principle) all entities in the industry and service sectors in 2013 9,14 , but 6 revised historical coal use for all years after 2000 upwards considerably (Figure 2). Total coal-derived 7 energy use for 2013 was revised up 13.6% (to 2.81 billion tonnes of coal equivalents (Gtce)) 15,16 -8 implying that 2014 coal-related CO 2 emissions in fact were much higher than existing estimates for 2013 9 -and every year since 2005 was revised up by 12-14%. Petroleum and natural gas saw only relatively 10 minor revisions (0%-1.7% and 0.7%-3.4%, respectively). The cumulative addition for 2000-2013 is 11 2.86 Gtce (+9.5%) from coal, 58.9 Mtce (+0.7%) fr...

show abstract

“…We use firm level panel data of Chinese manufacturing firms in 1998–2007, a data set that represents approximately 90 per cent of China's industrial output and 71 per cent of industrial employment as of 2004 (Brandt et al. ). The data comes from annual firm reports collected by the National Bureau of Statistics (NBS) and includes rich economic, financial, and ownership characteristics of each surveyed firm, as well as the key labour outcome variables that are central to our analysis.…”

Section: Research Data and Identification Approachmentioning

confidence: 99%

The Response of State‐owned Enterprises to Import Competition: Evidence From Chinese Manufacturing Firms

Duanmu

Pittman

2019

Ann Public Coop Econ

View full text Add to dashboard Cite

We examine the wage and employment impact on Chinese firms of an increase in import competition associated with China's WTO accession in December 2001, with an emphasis on state-owned enterprises (SOEs). We find that both wage and employment are negatively impacted by an increase in import competition, but firms with high state ownership cut employment less and reduce wages more than their private counterparts, suggesting that they prioritize the protection of employment over that of wages. This finding supports the notion that SOEs may have 'multitask' responsibilities in terms of protecting employment as well as achieving efficiency. We also find that firms with higher capital intensity reduce their wages less but cut employment more in response to intensified import competition. This provides empirical support for the efficiency wage theory.

show abstract

Challenges of working with the Chinese NBS firm-level data

Cited by 220 publications

References 17 publications

The Sources of Capital Misallocation

The Sources of Capital Misallocation

Uncertainties around reductions in China’s coal use and CO2 emissions

The Response of State‐owned Enterprises to Import Competition: Evidence From Chinese Manufacturing Firms

Contact Info

Product

Resources

About