China Economic Review 44 (2017) 271–281

Contents lists available at ScienceDirect

China Economic Review

journal homepage: www.elsevier.com/locate/chieco

Government intervention in the capital allocation process: Excess MARK

employment as an IPO selection rule in China
Anders C. Johanssona,⁎, Danglun Luob, Johanna Ricknec,d, Wei Zhenge
a
Stockholm School of Economics, Sweden
b
Lingnan College, Sun Yat-sen University, China
c
SOFI, Stockholm University, Sweden
d
Research Institute for Industrial Economics, Stockholm, Sweden
e
Shenzhen Graduate School of Harbin Institute of Technology, Shenzhen, China

AR TI CLE I NF O AB S T R A CT

JEL classifications: We study the role of excessive employment as a selection criterion for initial public offerings
D72 (IPOs) in China. Using a large dataset of firms that are eligible for a public offering, we find that
E24 firms' that have more excess employment – that is, firms that hire too many people – are more
G32 likely to be selected for an IPO. This correlation is stronger for the private sector than for the state
G34
sector, suggesting that stock market capital is used to direct capital flows to private firms that
Keywords: comply with politicians' preferred labor practices. A third set of results corroborates the
Local government inefficiency of this selection rule by showing that firms with more excess labor underperform
Initial public offering
after the IPO. We conclude that a political system known for its interventionistic government
Labor market
policies uses its influence over the stock market to signal preferred employment practices.
Excess labor
Capital allocation
China

1. Introduction

A well-functioning capital market is a cornerstone of economic development. Over the past decades, reforms to build external
capital markets have been commonplace in former socialist states and in emerging economies. China established their new stock
exchanges in the early 1990s, and its stock market has now grown to become the second largest in the world, both by trading volume
and market capitalization. As recently argued by Carpenter, Lu, and Whitelaw (2015), the future development of how well this
market allocates capital across investment opportunities will be an important determinant of global economic growth in the coming
decades.
A fundamental difference between the Chinese stock market and other major stock markets around the world is that the selection
process for IPOs is merit-based (Allen & Shen, 2010; Li & Zhou, 2015). This means that – despite decades of reforms that have made
the regulations more market based – the system is still one of case-by-case evaluation that is under close government control. Party-
state officials are closely involved in the process of helping firms apply for the IPO (Piotroski & Zhang, 2014), as well as in the
approval process of such applications (Hu & Liu, 2011; Li & Zhou, 2015). Francis, Hasan, and Sun (2009) and Li and Zhou (2015) find
that political connections help firms to preferential treatment in the IPO process, such as higher offering prices, lower underpricing,
and lower fixed costs during the IPO-process.

⁎
Corresponding author at: Stockholm School of Economics, P.O. Box 6501, SE-113 83 Stockholm, Sweden.
E-mail address: anders.johansson@hhs.se (A.C. Johansson).

http://dx.doi.org/10.1016/j.chieco.2017.05.002
Received 18 May 2016; Received in revised form 5 April 2017; Accepted 1 May 2017
Available online 04 May 2017
1043-951X/ © 2017 Elsevier Inc. All rights reserved.
A.C. Johansson et al. China Economic Review 44 (2017) 271–281

In this paper, we develop and test novel predictions that link the capital allocation process to politicians' political preferences. We
draw on a theoretical literature that has offered the important insight that politicians trade off employment creation for industrial
efficiency. Politicians prefer that firms employ more people than what would be needed to maximize their profit. To convince firms to
take on such excess labor, they receive financial transfers from politicians (e.g. Schleifer & Vishny, 1994; Boycko, Schleifer, & Vishny,
1996; Kornai, Maskin, & Roland, 2003; Röller & Zhang, 2005). Applying this basic insight to the stock market, we expect that firms
with more excess labor are more likely to be selected for an IPO. Following the same theories, we also expect that these transfers are
more important for firms in the private sector where, unlike for state-owned firms, managers can choose a smaller workforce unless
they receive compensation.
Using a large dataset of firms that are eligible for a public offering, we find support for capital (mis)allocation based on levels of
excess labor in firms. Excess labor is positively correlated with the probability of an IPO. As expected, this correlation is also stronger
for the private sector than for the state sector. This finding, that privately controlled firms are incentivized to create employment in
exchange for capital, sheds light on a novel way in which China's political system influences the behavior of privately controlled
firms.
By definition, the presence of redundant workers at a firm will hamper productivity. It is thus not unexpected that we can
corroborate the fact that firms with more excess labor underperform compared to other firms after the IPO. In addition, and although
the time truncation in our data does not allow a thorough analysis, it appears that firms maintain similar high levels of excess labor
decline in the post-IPO period.1 These results add gravitas to our findings by showing that the preference for the contribution to local
employment in the IPO selection process is conducive to an inefficient capital allocation via the domestic stock market.
In addition to the nascent literature on the political economy of China's stock market mentioned earlier, our study contributes to
several other strands of literature. First, our results extend the literature on policy burdens. Previous studies on public enterprises,
perhaps especially in transition economies such as China, have analyzed how governments have used SOEs to pursue non-financial
objectives (e.g. Lin, Cai, & Li, 1998; Lin & Tan, 1999; Lin & Li, 2008). Employment is a typical policy burden that SOEs must face.2 We
extend this literature by showing that governments may also force a policy burden such as promoting local employment upon
privately controlled firms in an indirect way, namely by offering a carrot in the form of access to finance through the equity market.
Second, we add to the understanding of how governments intervene in the capital allocation process and the economic
consequences of such interventions. There is a large literature on how government participation in finance is based on politicians'
desire to control investment by firms for motives that are either political (e.g. Schleifer & Vishny, 1994; La Porta, Lopez-de-
Silanes, & Shleifer, 2002) or social, i.e. an expression of a so-called benevolent government (see Shleifer (1998) for a detailed
discussion on this literature). In the Chinese case, it has previously been shown that SOEs with political relationships benefit from
access to bank loans and government subsidies (e.g. Brandt & Li, 2003; Tsai, 2007; Feng, Johansson, & Zhang, 2015;
Johansson & Feng, 2016). We add to this literature by showing how a specific objective, employment creation, influences the
capital allocation via a new channel, the stock market. We also show that this selection principle applies to private firms, a sector for
which we know less about how political concerns shape firm benefits. In addition, we add to the literature on negative effects of
government intervention, which have previously highlighted in studies on issues such as inefficiencies among state-controlled firms
and corresponding benefits of privatization (e.g., Megginson, Nash, & Randenborgh, 1994; Barberis, Boycko, Shleifer, & Tsukanova,
1996; Lopez-de-Silanes, Shleifer, & Vishny, 1997; Frydman, Gray, Hessel, & Rapaczynski, 1999; La Porta & Lopez-de-Silanes, 1999;
Chen, Sun, Tang, & Wu, 2011). Our empirical findings are in line with the argument that government intervention results in
inefficiencies.
Third and last, our work adds to the literature on IPOs and post-IPO performance. Several studies have analyzed post-IPO
performance and its potential determinants (e.g. Ritter, 1991; Jain & Kini, 1994; Brav & Gompers, 1997; Carter, Dark, & Singh, 1998;
Ritter & Welch, 2002; Kim, Kitsabunnarat, & Nofsinger, 2004). Feng, Johansson, and Zhang (2014) show that privately controlled
firms characterized by political ties in China tend to outperform other privately controlled firms. On the other hand, Fan, Wong, and
Zhang (2007) find that if all firms that have gone public in China are included in the analysis, firms with political ties tend to
underperform after the IPO. In a related paper, Cao, Ding, and Zhang (2016) analyze how different forms of social capital in the form
of political connections, external investors, and intragroup related-party transactions, are associated with post-IPO performance.
Wang, Xu, and Zhu (2004) link post-IPO performance in China to ownership structure. We add to this literature by focusing on
potential political motives behind IPO selection and the subsequent economic effects of a selection principle based on such motives.
Our results suggest that firms characterized by excess labor exhibit inferior post-IPO performance. The fact that this holds true for
privately controlled firms signals the importance of understanding of how firms are selected for China's stock market.
The rest of the paper is organized as follows: Section 2 discusses the institutional background and develops the working
hypotheses for the analysis. The focus is primarily on the IPO process in China and the potential role of political priorities for firms'
employment practices in that capital allocation process. In Section 3, we introduce the data and the key explanatory variable: excess
labor. Section 4 then presents and discusses the initial empirical results, focusing on the relationship between firm employment and
IPO selection. Section 5 addresses the economic implications of the relationship between firm employment and IPO selection by
analyzing post-IPO firm performance, and Section 6 concludes.

1
Sun and Liu (2012) argue that levels of excess labor should increase after an IPO when the capital injection from the stock market listing makes gives the firm a
greater financial capability to meet politicians' preferred employment practices.
2
In a related study, Huyghebaert and Wang (2012) find that having more board directors who are affiliated with the dominant shareholder is associated with more
labor redundancy in state-controlled firms. This suggests that state influence drives excess labor in Chinese SOEs.

272
A.C. Johansson et al. China Economic Review 44 (2017) 271–281

2. Institutional setting and hypothesis development

A fundamental difference between China's stock markets and other major stock markets in the world is the use of a merit-based
IPO system (Allen & Shen, 2010; Li & Zhou, 2015). Rather than using established criteria that firms need to meet in order to carry out
an IPO, the government controls the accessibility of the stock market for individual firms. Applications are judged by their merit on a
case-by-case basis. Li and Zhou (2015) observe that this feature has been highly resilient to system reforms. They note that “over the
past two decades, specific rules and regulations regarding share issuing in China have been frequently revised, [but] the fundamental
feature of China's share going public process, however, has remained largely unchanged, that is, under the merit-review regulatory
regime, the going public process remains under close control of the state”. Thus, despite reforms, the government remains in charge.
The main state organ that controls the IPO process is the China Securities Regulatory Commission (CSRC).3 It is an agency that
falls under the supervision of the central government. Historically, the IPO decision was based on both commercial and political
considerations. During the early 1990s, that is, before the sample period of data used in our empirical analysis, central authorities
determined the aggregate number of new IPOs that should be made available every year. The quota was then allocated across
ministries and provinces. Local authorities of security regulations invited prospective companies to request an IPO. The local
authorities then selected firms for IPOs based on both commercial considerations (profitability) and political objectives (social
development, employment, etc.).
In 1999, the introduction of the new China Securities Law meant that the quota system was, at least formally, abolished. The
“Rules in IPO and Listing on Stock Exchanges” was promulgated by the CSRC in 2006 and includes several criteria that must be met
before a firm can go public. These criteria require i) three consecutive years of net profit, ii) cumulative revenues three years prior to
the IPO and amounting to at least 300 million renminbi (RMB), iii) intangible assets not exceeding 20% of total assets, and iv) net
assets of at least 30 million RMB the year before the IPO. The Public Offering Review Committee (PORC) within the CRSC is
composed of 22 experts that review all applications for an IPO, making sure that they adhere to all laws and regulations related to
securities (Li & Zhou, 2015). Members of the PORC include both CSRC officials and external experts. In 2005, the CSRC took reforms
one step further with a revised Securities Law under which the CSRC no longer has to approve IPO prices.
Recent research has uncovered political influences over the merit-based IPO system. Using data prior to the 2005 pricing reform,
Francis et al. (2009) show that a firm with political connections could obtain higher IPO prices, a lower degree of IPO underpricing,
and lower IPO-related costs. Two other studies document how political connections shape the probability that a firm goes public in
the first place. Hu and Liu (2011) find that connections are helpful for private, but not state-owned firms. Assuming that all state-
owned firms are connected, Li and Zhou (2015) focus on private firms, finding that political connections are linked to a larger IPO
probability and several types of preferential treatment by the regulatory authorities. In these papers, political connections are defined
as having top officers who hold prominent positions in the national government, individuals who often hold additional political
positions at sub-national levels of government.
Another, and quite large, academic literature has considered political influences over capital allocation in a broader perspective
than just the stock market. In particular, capital allocation across companies is theoretically linked to politicians' preferences, which
are often modeled as twofold: efficiency and employment. Kornai et al. (2003) explain this in the context of “soft budget constraints”
of firms in socialist countries, i.e. they are insulated from bankruptcy by transfers from the government. Reviewing the substantial
literature on this practice the main motivation behind these transfers is given as: “On the one hand, the government wishes the
enterprise to earn a profit, because this enhances efficiency and provides a source of revenue. On the other hand, the government is
concerned that allowing a loss-making enterprise to fail will cause many workers to become redundant, thereby contributing to social
unrest and political tension” (Kornai et al., 2003, 1098).
One strand of research has modeled how a stylized politician directs financial flows to firms that provide so-called excess labor
(Schleifer & Vishny, 1994; Boycko et al., 1996; Röller & Zhang, 2005). In these models, the policy make, again, prefers firms to be as
productive (and profit-making) as possible, but also place a high value on creating employment and preventing unemployment.4 As
explained in the paper's introduction, excess labor refers specifically to a level of employment that is too high to be profit maximizing.
In order for firms to sacrifice profits in by offering this excessive employment level, the politician repays them with a financial
transfer of some sort.
In the context of the Chinese economy in particular, the imposition of the government's preferred labor practices, mainly to
employ redundant workers, in a way that lowers firm profits is often referred to as a (social) policy burden on firms (e.g. Lin et al.,
1998, Röller & Zhang, 2005; Lin & Li, 2008; Sun & Liu, 20125). The term derives from politicians' use of firms to deliver social policies,
for example employment security and social services (Dong & Putterman, 2003). As we will return to below, the size of these policy
burdens may have decreased in the transition period, but they are still expected to be substantial due to the fundamental fact that
employment remains the government's key concern (e.g. Lin et al., 1998; Lin & Tan, 1999; Lin & Li, 2008).
There are reasons to believe that the capital misallocation process that links capital to excess labor is made more severe by
another set of incentives facing Chinese politicians, namely from the career-promotion system. This system is perhaps most relevant
for sub-national level politicians,6 who are also intimately involved in the IPO process, albeit at its' earlier stages. Piotroski and Zhang

3
For more details on the history of China's stock markets during the economic reforms, see Johansson and Ljungwall (2009).
4
While the other papers focus on socialist countries, where regime stability is usually at the core of the employment preference, Schleifer and Vishny (1994) discuss
democracies. In this context, the argument is instead that politicians cater to interest groups, particularly to labor unions, rather than to the median voter.
5
As expressed by Sun and Liu (2012, p. 125): “once the optimum employment level is exceeded, the employment scale forms into the policy burden of the firm.”
6
China has five levels of local government, where provinces is the layer immediately below the national government.

273
A.C. Johansson et al. China Economic Review 44 (2017) 271–281

(2014) observe that in addition to meeting the formal IPO criteria, local firms still need the support and approval of local politicians
and party committees to submit an application.
What are, then, the incentives that local politicians face? In a nutshell, to advance in the party-state structure, local party-state
officials compete against each other in “promotion tournaments” based on the performance of the regions under their control (for a
detailed description, see Xu, 2011). Although the precise criteria for upward mobility remain unknown to the research community,
some consensus has emerged. Economic growth is considered to be of considerable importance (Maskin, Qian, & Xu, 2000; Li & Zhou,
2005; Chen, Li, & Zhou, 2005) in addition to “soft” targets such as maintaining social stability (Edin, 2003; Landry, 2008).7
Notably, employment can boost growth numbers while at the same time increasing social stability, yielding dual contributions to
officials' promotion chances. Sun and Liu (2012) describe how local politicians at the province level strive to produce a batch of listed
companies as part of their development strategies for provincial capitals. These strategies form part of the politician's office
maximizing calculus specifically via the ambition of (excess) employment creation. Listed firms have i) expanded financial
capabilities to hire redundant workers and ii) are placed more tightly under political control so that their employment decisions can
be influenced.
Interestingly, another aspect of the career-promotion system may twist local politicians' incentive structure even further away
from efficient capital allocation. Government officials are evaluated every fifth year, and reshuffling commonly occurs in off-years as
well. This rewards policies that favor short-term growth over long term development (e.g. Xu, 2011).8 So, while promoting the most
productive firms for an IPO may be efficient for the economy in the long run, picking firms that offer immediate employment benefits
may more conducive to immediate political benefits.
Pulling together these strands of research about political incentives, financial flows and excess employment, it seems reasonable
that excess employment could form part of the capital allocation process via the stock market. While the Chinese stock markets are
continuously becoming more market-oriented, government involvement remains high. The trade-off between efficiency and political
objectives, particularly in the area of employment, is likely to arise in this area, both as a consequence of general objective functions
of politicians, and because of the particular incentive structures embedded in the promotion tournament system. IPOs can function as
a financial transfer in exchange for excess employment. The use of this selection rule can also fill the function of signalling the value
that politicians place on such employment practices to firms that might desire a listing in the future. We therefore expect that:

Hypothesis 1. Firm with more excess labor are more likely to be accepted for an IPO.
Research on capital allocation and excess employment often considers the distinction of ownership sectors, in particular between
state- and private-owned firms. Under the planned economy, all firms (then called work units) where state-owned and provided
employment and other social services to the (urban) population. The state's commitment to employing the population, combined with
its' industrial strategy of promoting heavy industry with relatively small labor needs, created a situation where SOEs were assigned
redundant workers (e.g. Lin et al., 1998; Dong & Putterman, 2003). Correspondingly, the release of (some of) this excess labor via the
process of SOE reforms in recent years has been an important source of industrial productivity growth between 1998 and 2007
(Hsieh & Song, 2016).
The policy burden of excess labor is, thus, generally assumed to be larger for state-owned firms (e.g., Boycko et al., 1996; Kornai
et al., 2003; Dong & Putterman, 2003; Xu, Zhu, & Lin, 2005; Röller & Zhang, 2005). Theoretically, this is linked – of course – to the
government's direct control of management decisions (e.g., Boycko et al., 1996). Nor is the government's control expected to be
diminished after the stock market listing of SOEs (Xia & Fang, 2005; Lin & Milhaupt, 2013). But the burden of excess labor is also
expected fall on private, or privatized, firms. Lin et al. (1998, 429f) note that “because the state is accountable for the policy burdens,
the [privatized] enterprises, profitable or not, will bargain with the state for ex ante policy favors, such as access to low-interest loans,
tax reductions, tariff protection, legal monopolies, and so on, in order to compensate for the burdens. The state will be in a difficult
position to resist the enterprises' pressures for such favors”.
The key difference between the sectors is that the managers of private firms will not over-employ unless offered financial
incentives (Boycko et al., 1996; Röller & Zhang, 2005). Financial incentives are therefore offered to private firms to convince
efficiency-oriented managers to make inefficient decisions. In addition to this direct favor-transfer between politicians and private
firms, where excess labor is traded for stock market funding, the signalling effect could also be particularly relevant for the private
sector. Because the government lacks influence over the labor practices of private firms, a signal of preferential access to finance in
the form of an IPO could influence a pool of firms in the local economy that the government has less influence over. We therefore
expect that:

Hypothesis 2. The relationship between excess labor and the likelihood of an IPO is stronger for privately controlled firms than for
state controlled firms.
Finally, excess labor may have adverse economic effects. This is indeed expected from the very definition of the concept, where a
firm is hiring a sub-optimally large workforce. This fact is underscored in the trade-off between productivity and political objectives
in the theory presented above. Recent empirical research has also linked labor misallocation to lower productivity in the Chinese

7
Having personal connections to officials at higher levels also helps, and function as a complement to economic growth, in particular at the highest tiers of the
promotion ladder (argument and literature summarized by Landry, Lu, & Duan, 2016).
8
For example, Jia (2017) shows that promotion incentives are linked to more polluting development strategies at the local level, and Guo (2009) discusses how
promotion incentives are conducive to debt-financed, large development projects known as “political achievement projects”.

274
A.C. Johansson et al. China Economic Review 44 (2017) 271–281

industrial sector compared to other countries (Hsieh & Klenow, 2009). In addition, the policy burden of excess labor has been
pronounced as an important reason for the state sector's lower productivity (Xu et al., 2005; Hsieh & Song, 2016). Sun and Liu (2012)
discuss two specific channels through which excess labor can impact on productivity. First, high labor costs can force managers to
lower wages, making workers disgruntled and management more difficult. Second, excess employment can become a “perfect excuse”
for less competent managers, making performance evaluations noisier and, in turn, weakening corporate governance (Sun & Liu,
2012, p. 128). In addition, Bai, Yang, Xue, and Jin (2010) find that excess labor is from the obstruction it brings to compensation-
based incentive mechanisms, resulting in an agency cost for management and less focus on firm performance.
To sum up, excess labor may be required for the firm to be selected for an IPO. If the firm takes on excess employment as a signal
to the local government, taking on this additional cost could help explain subsequent underperformance. We therefore hypothesize as
follows:
Hypothesis 3. Firms with more excess labor underperform after the IPO.

3. Research design

3.1. Data sample

The focus of this study is on firms that go public in China. Our analysis on potential determinants of the IPO event means that we
first need to identify a set of eligible firms. Our identification of these firms is closely related to the literature on the decision to go
public (e.g. Pagano, Panetta, & Zingales, 1998). Similar to Piotroski and Zhang (2014), we use a database of unlisted companies that
is compiled and managed by China's National Bureau of Statistics (NBS). Chinese regulations state that all companies with sales over
5 million renminbi are required to report their financials to the NBS using a standardized system. As pointed out by Piotroski and
Zhang (2014), the dataset provided by the NBS covers around 90% of the gross industrial productivity in China. Furthermore, and
most importantly, we can assume that all the firms that are eligible for a public offering are included in the sample. As outlined above,
having revenues above 300 million RMB over a three-year period is one of the formal criteria for an IPO. The NBS data set covers the
period 2001–2008 and we therefore must restrict the analysis to this period.
To compile a data set of eligible firms, we first identify firms in the NBS dataset that meet the CSRC's criterion mentioned earlier.
This process utilizes data for the three previous years, meaning that a firm can enter only when we have three years of previous
observations to assess. From this sample, we only keep observations for firms that meet the criterion, i.e. that are formally eligible for
an IPO. Note that control variables for the analysis can still be calculated based on years of data when the firm was not eligible.
Having compiled a sample of all eligible firms, we then identify all firms that went public during the period 2001–2008 by using
the China Securities Market and Accounting Research (CSMAR) IPO database. Pre-IPO firm data is obtained from the NBS database,
while post-IPO firm data for all listed firms is obtained from CSMAR's database.
The firm's ownership category is central to our empirical analysis. We use the following procedure to separate firms based on
whether they are owned by the state or not. The NBS database provides the firm's registered ownership type, which we divide into the
categories of state- or private-owned. For firms that do not belong to either of these groups we use paid-in capital to define ownership
(e.g. for the categories of collective-owned firms, shareholding joint ventures, collectively joint venture, limited liability corporation,
or shareholding company). We can define any company in these categories as owned by the state if the state owns more than 50% of
the shares at the time of the offering (Piotroski & Zhang, 2014). Moreover, we also look at both code and name of the firm in the
database. This is because these are not matched uniquely. The same code could for example refer to different names in different years.
Similarly, the same name could refer to different codes in different years. We therefore first sort the data by code and then check if a
code matches uniquely with a firm name. If this is not the case, we make use of other firm information such as location and legal
person to identify the firm in question. This approach minimizes the risk of mistakes and allows us to obtain a sample that is
approximately 10% larger than the one used by Piotroski and Zhang (2014).

3.2. Measurement of excess labor

The key dependent variable is our measure of firms' level of excess labor. This variable is estimated in an expectation model (e.g.
Zeng & Chen, 2006), where a first step in the estimation procedure relates the firm's employment to a number of determinants of firm
labor using the specification.
laborit = β1 sizeit + β2 tangibilityit + β3 growthit + ∝ind + ∝t εit (1)
Here, laborit is the number of employees over total sales (in ten thousands) for firm i at time t. As determinants, we use standard
explanatory variables for firm employment, namely: size, the logarithm of total assets; tangibility, asset tangibility calculated as fixed
assets over total assets; growth, measured as the sales growth rate. All these variables are three-year averages. We also include
industry and year dummies. In a second step, the estimated values for the coefficients in Eq. (1) are used to obtain the predicted value
of labor for each firm in each year:
laborit = b1 sizeit + b2 tangibilityit + b3 growthit + ∝ind + ∝t (2)
The intuition is that these predicted values can be interpreted as the normal employment level for a given industry-year-
characteristic cell. Our measurement of excess labor is the difference between a firm's actual employment level and this “normal”

275
A.C. Johansson et al. China Economic Review 44 (2017) 271–281

Table 1
Descriptive statistics.
This table presents the number of firms that are eligible for an initial public offering (IPO) as well as the number of firms that actually carried out an IPO in China
over the sample period 2001–2008. The sample is also divided into two subsamples with state- and non-state-owned firms, respectively.

Year Total number of Number of non-state- Number of state- Number of firms Number of non-state-owned Number of state-owned
firms owned firms owned firms going public firms going public firms going public

2001 1779 1190 589 57 16 41

2002 2113 1545 568 51 15 36
2003 2600 1967 633 46 20 26
2004 3035 2420 615 79 50 29
2005 3619 2984 635 12 7 5
2006 4490 3823 667 42 28 14
2007 6198 5391 807 88 68 20
2008 7611 6783 828 57 47 10
Total 31,445 26,103 5342 432 251 181

level:
excess laborit = laborit − laborit (3)

4. Excess labor as an IPO selection rule

With the data and variables at hand, this section presents descriptive statistics and describes our empirical approach for testing
the first two theoretical hypotheses. We then present the results for relating the probability of an IPO to excess labor, both in the total
sample (Hypothesis 1) and in separate sample of privately controlled and state owned firms (Hypothesis 2).

4.1. Descriptive statistics

Table 1 presents descriptive statistics for firms that are identified as being eligible to go public during the period 2001–2008. We
identify 31,445 firms that fulfill the formal criteria for a public offering during the sample period. Out of these, a total of 26,103 firms
are privately controlled, while 5342 are owned by the state. It has been shown that the private sector has grown considerably and that
it is one of the main drivers of economic growth in China during the last decades (e.g. Allen, Qian, & Qian, 2005; Dougherty,
Herd, & He, 2007). Our descriptive statistics reflect this transformation, with an increasing share of the larger firms in China being
owned by non-state entities. While most firms in the sample are privately controlled, SOEs are more likely to be chosen for an IPO.
432 firms carried out an IPO during the sample period, representing 1.4% of the total sample, among which 251 were privately
controlled (representing 1.0%) and 181 were state-owned (representing 3.4%). This is in line with the argument that SOEs receive
preferential treatment and that the Chinese banking system and stock market have been used by the government to finance firms
controlled by the state (e.g. Walter & Howie, 2011).
Table 2 presents an overview of excess labor in listed and unlisted firms. As expected given the share of these firms in the total
sample, the mean of excess labor for unlisted firms is zero for all years across the sample. For listed firms, on the other hand, the
picture is different. Excess labor is positive for every year between 2001 and 2008. When we test the difference for the mean and
median between the two samples, we find that it is significantly different for all years (except for the difference in the mean for 2008).

Table 2
Excess labor across subsamples based on stock market listing status.
This table presents the mean and median of excess labor over the sample period 2001–2008. The sample is divided into listed and unlisted firms respectively.
Differences in mean and median are also presented.

Excess Labor Listed firms Unlisted firms Difference

Obs. mean median Obs. mean median mean median

2001 57 0.011 0.003 1722 0.000 − 0.009 0.012⁎⁎⁎ 0.012⁎⁎

2002 51 0.012 0.009 2062 0.000 − 0.009 0.012⁎⁎⁎ 0.018⁎⁎⁎
2003 46 0.007 0.001 2554 0.000 − 0.008 0.007⁎ 0.008⁎⁎
2004 79 0.005 − 0.003 2956 0.000 − 0.007 0.005⁎⁎ 0.004⁎⁎⁎
2005 12 0.019 0.018 3607 0.000 − 0.007 0.019⁎⁎⁎ 0.024⁎⁎⁎
2006 42 0.011 0.002 4448 0.000 − 0.006 0.011⁎⁎⁎ 0.008⁎⁎
2007 88 0.005 0.001 6110 0.000 − 0.006 0.005⁎⁎ 0.007⁎⁎⁎
2008 57 0.003 0 7554 0.000 − 0.005 0.003 0.004⁎⁎⁎
total 432 0.007 0.001 31,013 0.000 − 0.006 0.007⁎⁎⁎ 0.007⁎⁎⁎

⁎⁎⁎
Denotes statistical significance at the 1% level.
⁎⁎
Denotes statistical significance at the 5% level.
⁎
Denotes statistical significance at the 10% level.

276
A.C. Johansson et al. China Economic Review 44 (2017) 271–281

Table 3
Excess labor across various subsamples.
This table presents the mean and median of excess labor over the sample period 2001–2008. The sample is divided into listed and unlisted firms. The sample is also
divided into subsamples for state- and non-state-owned firms, as well as for firms located in regions with high and low employment pressure. Employment pressure is
measured by the three-year average growth rate of unemployment at the provincial level. Differences in mean and median are also presented.

Listed firms Unlisted firms Difference

Mean Median Mean Median Mean Median

Non-state-owned 0.0039 −0.0005 − 0.0015 − 0.0069 0.0054⁎⁎⁎ 0.0064⁎⁎⁎

State-owned difference 0.0120 0.0042 0.0068 − 0.0014 0.0053⁎⁎⁎ 0.0056⁎⁎⁎
−0.0082⁎⁎⁎ −0.0047⁎ − 0.0082⁎⁎⁎ − 0.0055⁎⁎⁎
High employment pressure 0.0090 0.0016 0.0006 − 0.0062 0.0084⁎⁎⁎ 0.0078⁎⁎⁎
Low employment pressure difference 0.0041 0.0007 − 0.0008 − 0.0061 0.0048⁎⁎⁎ 0.0068⁎⁎⁎
0.0049⁎⁎ 0.0009 0.0013⁎⁎⁎ − 0.0001

⁎⁎⁎
Denotes statistical significance at the 1% level.
⁎⁎
Denotes statistical significance at the 5% level.
⁎
Denotes statistical significance at the 10% level.

These preliminary findings lend support to the conjecture that access to financing via public offerings is directed toward firms with
more excess labor.
Table 3 again shows the differences in excess labor across listed and unlisted firms and across privately and state owned firms,
respectively. We also divide the sample into subsamples depending on ownership. Tests for differences in the mean and median also
show that – as expected – firms controlled by the state exhibit significantly more excess labor among than privately controlled firms.
This is true both among listed and unlisted firms.
We also compare levels of excess labor by the level of provincial unemployment from China Statistical Yearbook (various years).
The correlations show that firms located in regions characterized by high unemployment exhibit significantly more excess labor on
average, and this holds for both unlisted and listed firms. Moreover, tests for differences in the mean and median between unlisted
and listed firms show that listed firms exhibit significantly more excess labor where unemployment is higher.

4.2. Baseline results

Our first hypothesis is that the IPO event is influenced by the firm's contribution to the local labor market (excess labor). To
analyze this hypothesis, we estimate a logit regression model for the relationship between excess labor on the IPO event. The model
estimates the probability that an eligible unlisted firm will carry out an IPO in a year t and is specified as:
⎛ p (ipo) ⎞
ln ⎜ ⎟ = βexlaborit + X′it γ + ∝ind + ∝t + εit
⎝ 1 − p (ipo) ⎠ (4)
′
In addition to the key explanatory variable excess labor (excess labor) introduced in Section 3.2, Xi is a vector of additional
potential determinants of the IPO event: pgdp, the provincial GDP per capita; roa, return on assets; fcf, free cash flow; growth,
measured as the sales growth rate; investments, calculated as investment over total assets; soe; a dummy which is equal to 1 if the firm
is owned by the state and 0 otherwise; leverage, calculated as total liabilities over total assets; tangibility, measured as fixed assets over
total assets; size, defined as the logarithm of total assets. We also include two controls for other politicized capital transfers that could
help the firm in the IPO application procedure, namely subsidies, the amount of government subsidies; and interest, a measure of the
firm's interest payments divided by total sales. Industry (∝ind) dummies at the 1-digit industry level, and year (∝t) dummies are also
included but not reported. See Table A1 for definitions of all variables used in the empirical analysis. We cluster the standard errors at
firm level.
Column (1) in Table 4 presents coefficients from the estimation of Eq. (4). The signs of the control variables come out as
expected.9 More importantly, we find a significant positive relationship between excess labor and the likelihood of going public for
the whole sample. This finding confirms our first hypothesis: firms with more excess labor are more likely to go public.
We can also examine the explanatory power of the economic variables relative to the political variables. Starting from a
specification that only includes the fixed effects, this regression explains 4% of the variance in the outcome variable (not reported).
Adding economic variables, that is, all variables except for excess labor, the state-ownership dummy, and the subsidies and interest
payment variables, takes the explained variance to 7.9%. Removing the economic controls and adding instead the political variables
has a similar effect, doubling the explained variance to 8.1%. In sum, the political variables appear to be equally predictive as the
economic variables.
Temporal variation in market reforms could be a potential source of variation in our results. Split sample analysis for each year in
our data shows that, if anything, excess labor has become a slightly stronger correlate with the IPO event over time (not reported). We
return to this observation in the paper's conclusions.

9
Results are not affected by also controlling for the number of previous IPOs in each province.

277
A.C. Johansson et al. China Economic Review 44 (2017) 271–281

Table 4
Excess labor and the IPO event.
This table presents results from the estimation of a logit model for IPO decision for a sample of firms that are eligible for a public offering during the
period 2001–2008. The dependent variable is the likelihood that a firm carries out an IPO in year t. Excess labor is calculated by first estimating the
predicted value of labor and then subtracting this from the firm's actual value of labor, where value of labor is equal to number of employees over total
sales in ten thousands. The final variable is the standardized (z-score) of these deviations. pgdp is the provincial GDP per capita, roa (return on assets) is
calculated as earnings before interest and taxes over total assets, fcf is the free cash flow, growth is the growth in sales, investment is investments over total
assets, soe is a dummy which is equal to 1 if the firm is owned by the state and 0 otherwise, leverage is total liabilities over total assets, tangibility is fixed
assets over total assets, size is the logarithm of total assets. Industry and year dummies are included but not reported.

(1) (2) (3)

Full sample Private sector State sector

excess labor 0.141⁎⁎⁎ 0.228⁎⁎⁎ 0.0287

(0.0445) (0.0611) (0.0663)
pgdp − 0.248⁎⁎ − 0.0455 −0.599⁎⁎⁎
(0.111) (0.148) (0.179)
roa 4.132⁎⁎⁎ 4.266⁎⁎⁎ 4.799⁎⁎⁎
(0.751) (0.956) (1.249)
fcf − 0.270 − 0.268 −0.0268
(0.284) (0.345) (0.537)
growth − 0.332⁎⁎ − 0.265 −0.536⁎
(0.160) (0.197) (0.286)
investment 2.154⁎⁎⁎ 2.048⁎⁎⁎ 2.603⁎⁎⁎
(0.443) (0.572) (0.743)
leverage 2.684⁎⁎⁎ 3.130⁎⁎⁎ 1.889⁎⁎⁎
(0.363) (0.482) (0.579)
tangibility − 0.587⁎ − 0.757⁎ −0.465
(0.307) (0.418) (0.469)
size − 0.300⁎⁎⁎ − 0.226⁎⁎⁎ −0.401⁎⁎⁎
(0.0593) (0.0830) (0.0860)
subsidies 43.35⁎⁎⁎ 53.12⁎⁎⁎ 26.60⁎
(8.803) (11.08) (14.73)
interest payments 11.50⁎⁎⁎ 19.46⁎⁎⁎ 5.430
(3.300) (4.601) (4.803)
SOE 1.252⁎⁎⁎
(0.119)
Industry FE & Year FE yes yes yes
N 31,445 26,103 5342
pseudo R2 0.118 0.084 0.136

⁎⁎⁎
Denotes statistical significance at the 1% level.
⁎⁎
Denotes statistical significance at the 5% level.
⁎
Denotes statistical significance at the 10% level.

Finally, before turning to the next hypothesis, a possible robustness check is to test whether total firm employment is also
correlated with a larger IPO probability. Even if our theory specifically focuses on excess labor, for the obvious reason that this
variable trades off employment against efficiency, politicians may – according to a similar logic – give preference to large firms.
Indeed, replacing excess labor with the log of the number of employees gives a positive and significant result (not reported), which
could be further explored in future work.
So far, we have identified a significant and positive relationship between excess employment and IPO selection. However, we
have yet to shed light on whether all firms regardless of ownership type are treated the same way, or if IPO selection is specifically
targeting a certain group of firms. The findings in the previous section corroborated the argument that the government is giving
preferential treatment to firms under state ownership as they, all else being equal, are more likely to go public than privately
controlled firms.
As stated in Hypothesis 2, it is likely that excess labor is more important for privately controlled firms in their pursuit of going
public. To further our understanding of how IPO selection is used as a governance mechanism, i.e. as a signal to other firms of the
government's preferred labor practices, we therefore divide the sample into two groups of firms based on ownership. We then run the
regression in Eq. (4) on each of the two subsamples separately. The new results, which are presented in Columns (2) and (3) in
Table 4, help shed additional light on IPO selection as a governance mechanism. Focusing on the key explanatory variable, the
coefficient for excess labor is no longer significant in the sample of state-owned firms, and it is also smaller than the positive and
significant coefficient in the sample of private-owned firms. These findings suggest that excess labor enters the IPO selection process
for firms that are privately owned, possibly because these managers cannot be otherwise convinced to comply with the politician's
employment preference.

5. Economic implications: post-IPO firm performance

So far, we have seen that firms that contribute more to local social objectives, such as keeping unemployment down, are more

278
A.C. Johansson et al. China Economic Review 44 (2017) 271–281

likely to receive capital via the stock market. But how about the potential negative effects? As stated in Hypothesis 3, excess labor
should hurt firm performance. To shed light on this issue, we analyze the relationship between excess labor and post-IPO performance
using the regression model in Eq. (5) below. Consistent with the literature on post-IPO performance (e.g. Fan et al., 2007; Feng et al.,
2014), we use pre-IPO operating performance figures as benchmark when we evaluate firms' post-IPO performance. The change in
performance (change_peri) is calculated by subtracting the average of the annual performances three years prior to the IPO from the
average of the annual performances three years after the IPO. We use three performance measures: return on sales, industry-adjusted
return on sales, and earnings growth.10

changeperi = ∝0 + β1 excess labori + β2 tangibilityi + β3 liquidityi + β4 sizei + β5 leveragei + β6 soei + β7 Subsidiesi + β8 Interesti +
∝ind + ∝t + εit (5)
In addition to our main explanatory variable, we also include standard control variables (e.g. Fan et al., 2007; Feng et al., 2014)
plus the two controls for monetary transfers from the previous analysis, namely Subsidies and Interest Payments (as before, Appendix
Table A1 provides an overview of the variables): tangibility, which is fixed assets over total assets; liquidity, measured as current assets
over total assets; firm size, measured as the logarithm of total assets; leverage, measured as total liabilities over total assets; soe, a
dummy for state ownership; subsidies, the amount of government subsidies; and interest, the firm's interest payments divided by total
sales. Industry and year dummies are once more included but not reported. Before moving on to the empirical analysis, we also check
for the possibility that a firm is delisted shortly after its IPO, something that could potentially drive the results.11 It turns out that
delisting shortly after an IPO is rare during the period in question, which means that this does not represent a potential issue in the
analysis.
The results from the estimations for each of the three performance measures are presented in Table 5. Among the control
variables, only leverage has a significant effect on post-IPO performance. Its coefficient is significant positive, but only for the whole
sample and the subsample of privately controlled firms, suggesting that private firm that can obtain access to finance and increase
their leverage can perform better after going public. Focusing on excess labor, the results suggest that it plays an important role for
privately controlled firms. While the whole sample and the SOE subsample are not affected by excess labor, post-IPO performance of
privately controlled firms exhibit a significant negative relationship with excess labor. These results, which are evidence in support of
our third and final hypothesis, suggest that privately controlled firms must pay a price for contributing to social objectives such as
local employment. Privately controlled firms in China thus must balance the pros and cons of contribution to local social objectives as
part of their long-term strategy. Moreover, these findings indicate that contributions to local social objectives by private firms result
in inefficiencies in the capital allocation process.

6. Conclusion

This study provides evidence that firms' employment of redundant workers, so-called excess labor, enters the IPO decision process
in China. We also found that excess labor matters for the capital allocation to private firms, but not to state-owned firms, presumably
because private sector managers require transfers to engage in sub-optimal labor practices (e.g. Boycko et al., 1996). For state-owned
firms, they already face a policy burden that typically includes contributions to the local labor market, which means that the decision
of whether or not to grant them access to financing through the stock market is not dependent on their labor intensity.
Prior research has shown that firms that contribute to political (and social) objectives regularly receive benefits such as
preferential access to capital in the form of government subsidies and bank loans. Our study contributes to this literature by providing
evidence that access to financing through IPOs is associated with contribution to local employment.
Our findings shed new light on how the Chinese government expands the domestic stock market. While formal requirements that
focus on the financial soundness and growth prospects of firms are in place, informal requirements such as employment practices also
play an important role in the IPO decision process. Our findings on post-IPO performance further suggest that these informal
requirements have distortive effects on the capital allocation process in China. Excess labor is correlated with worse performance
after the IPO. This result contributes to our understanding of the relatively poor performance of China's two stock markets, i.e. that
the capital allocation process via the stock market takes account of excess labor and thereby sacrifices efficiency for the sake of
political preferences.
In the period that we study, 2001–2009, we saw, if anything, an increase over time in the importance of excess labor in the IPO
selection process. This conforms with an observation by Li and Zhou (2015), namely that market reforms have not removed the
selection out of the hands of politicians, even to the present day. Recent reforms also highlight the government's view that capital
allocation via the stock market continues to be the subject of political priorities, for example by the announcement of the IPO poverty
alleviation program in 2016 (Chen, 2016). These reforms run counter to the idea of a pure, market-based system, which could be
combined with other modes of regional redistribution.

10
It should be noted that Gormley and Matsa (2014) have argued that the use of industry-adjusted dependent variables produce inconsistent estimates. To alleviate
the risk of distorted inference based on our estimations, we therefore use two alternative measures for performance in addition to industry-adjusted return on sales.
11
We thank an anonymous reviewer for pointing this out.

279
A.C. Johansson et al. China Economic Review 44 (2017) 271–281

Table 5
Post-IPO firm performance.
This table presents results from various estimations of a cross-section regression model for post-IPO performance for firms that went public in China during the
period 2001–2008. The dependent variables are the change in the three-year average return on sales, change in the three-year average adjusted return on sales, and the
three-year average growth in earnings, respectively. excess labor is calculated by first estimating the predicted value of labor and then subtracting this from the firm's
actual value of labor, where value of labor is equal to number of employees over total sales in ten thousands. The final variable is the standardized (z-score) of these
deviations. tangibility is fixed assets over total assets, and liquidity is the current asset over total asset. size is the logarithm of total assets, leverage is total liabilities over
total assets, soe is a dummy which is equal to 1 if the firm is owned by the state and 0 otherwise.

Change in ROS Change in adjusted ROS Earnings growth

All Non-state State All Non-state State All Non-state State

⁎⁎ ⁎⁎ ⁎⁎
excess labor −0.006 − 0.016 0.001 − 0.007 −0.016 0.001 −0.035 − 0.127 0.008
(0.005) (0.008) (0.006) (0.005) (0.008) (0.006) (0.036) (0.058) (0.045)
tangibility 0.188 0.041 0.455⁎ 0.190 0.039 0.445⁎ 0.971 0.590 1.855
(0.141) (0.153) (0.239) (0.138) (0.156) (0.227) (1.000) (1.312) (1.497)
liquidity 0.116 − 0.034 0.383⁎ 0.108 −0.060 0.389⁎ 0.759 − 0.147 2.154
(0.126) (0.133) (0.221) (0.124) (0.135) (0.212) (0.931) (1.116) (1.519)
size −0.005 − 0.007 0.001 − 0.005 −0.011⁎ 0.002 −0.018 − 0.119 0.085
(0.006) (0.006) (0.009) (0.006) (0.006) (0.009) (0.048) (0.079) (0.060)
leverage 0.151⁎⁎⁎ 0.195⁎⁎⁎ 0.070 0.143⁎⁎⁎ 0.197⁎⁎⁎ 0.052 −0.103 0.122 −0.402
(0.053) (0.059) (0.099) (0.052) (0.060) (0.098) (0.341) (0.412) (0.654)
SOE -0.005 − 0.002 −0.028
(0.012) (0.012) (0.104)
subsidies −1.560⁎⁎⁎ − 2.522⁎⁎⁎ − 1.033⁎ − 1.255⁎⁎⁎ −2.038⁎⁎ − 0.869 −1.624 − 2.707 −0.569
(0.459) (0.804) (0.532) (0.457) (0.802) (0.526) (3.268) (5.541) (4.060)
interest payments −1.945⁎ − 2.759⁎ − 0.045 − 2.496⁎⁎ −3.041⁎⁎ − 0.474 −9.044 − 7.413 −2.558
(1.168) (1.490) (1.880) (1.153) (1.478) (1.892) (7.778) (10.119) (11.337)
Ind FE & year FE Yes Yes Yes Yes Yes Yes Yes Yes Yes
N 431 251 180 431 251 180 431 251 180
R2 0.099 0.174 0.110 0.089 0.149 0.123 0.040 0.079 0.127

⁎⁎⁎
Denotes statistical significance at the 1% level.
⁎⁎
Denotes statistical significance at the 5% level.
⁎
Denotes statistical significance at the 10% level.

Appendix A

Table A1
Variable definitions.

Variable Definition

ipo Dummy for firms succeed in going public; ipo = 1 if the firm carries out an IPO and 0 otherwise
excess labor Estimated as explained in Section 3.2
pgdp GDP per capita
roa Return on assets
fcf Free cash flow
growth Sales growth rate
investment Investments/total assets
soe Dummy for state ownership; soe = 1 if the firm is owned by the state and 0 otherwise
leverage Total liabilities/total assets
tangibility Fixed assets/total assets
liquidity Current assets/total assets
size Log(total assets)
subsidy government subsidies/total sales
interest interest payments/total sales

References

Allen, F., Qian, J., & Qian, M. (2005). Law, finance and economic growth in China. Journal of Financial Economics, 77, 57–116.
Allen, W., & Shen, H. (2010). Assessing China's top-down securities markets. In R. Morck, & J. Fan (Eds.), Capitalizing China. Chicago: University of Chicago Press.
Bai, Y., Yang, J., Xue, Y., & Jin, Y. (2010). The impact of excess employment on Chinese state-controlled enterprises: An empirical study. International Journal of
Sustainable Economy, 2, 32–58.
Barberis, N., Boycko, M., Shleifer, A., & Tsukanova, N. (1996). How does privatization work? Evidence from the Russian shops. Journal of Political Economy, 104,
764–790.

280
A.C. Johansson et al. China Economic Review 44 (2017) 271–281

Boycko, M., Schleifer, A., & Vishny, R. W. (1996). The Economic Journal, 106(435), 309–319.
Brandt, L., & Li, H. (2003). Bank discrimination in transition economies: Ideology, information or incentives? Journal of Comparative Economics, 31, 387–413.
Brav, A., & Gompers, P. A. (1997). Myth or reality? The long-run underperformance of initial public offerings: Evidence from venture and nonventure capital-backed
companies. Journal of Finance, 52, 1791–1821.
Cao, J. X., Ding, Y., & Zhang, H. (2016). Social capital, informal governance, and post-IPO firm performance: A study of Chinese entrepreneurial firms. Journal of
Business Ethics, 134, 529–551.
Carpenter, J. N., Lu, F., & Whitelaw, R. F. (2015). The real value of China's stock market. NBER working paper no. 20957.
Carter, R. B., Dark, F. H., & Singh, A. K. (1998). Underwriter reputation, initial returns, and the long-run performance of IPO stocks. Journal of Finance, 53, 285–311.
Chen, C.. CSRC's latest IPO plan: Public folly or sensible policy? South China Morning Post, September 15, 2016. (2016). Retrieved from http://www.scmp.com/business/
companies/article/2019702/csrcs-latest-ipo-plan-public-folly-or-sensible-policy (accessed March 28, 2017) .
Chen, S., Sun, Z., Tang, S., & Wu, D. (2011). Government intervention and investment efficiency: Evidence from China. Journal of Corporate Finance, 17, 259–271.
Chen, Y., Li, H., & Zhou, L. (2005). Relative performance evaluation and the turnover of provincial leaders in China. Economics Letters, 88, 421–425.
Dong, X.-Y., & Putterman, L. (2003). Soft budget constraints, social burdens, and labor redundancy in China's state industry. Journal of Comparative Economics, 31,
110–133.
Dougherty, S., Herd, R., & He, P. (2007). Has a private sector emerged in China's industry? Evidence from a quarter of a million Chinese firms. China Economic Review,
18, 309–334.
Edin, M. (2003). State capacity and local agent control in China: CCP cadre management from a township perspective. China Quarterly, 173, 35–52.
Fan, J. P. H., Wong, T. J., & Zhang, T. (2007). Politically connected CEOs, corporate governance, and post-IPO performance of China's newly partially privatized firms.
Journal of Financial Economics, 84, 330–357.
Feng, X., Johansson, A. C., & Zhang, T. (2014). Political participation and entrepreneurial initial public offerings. Journal of Comparative Economics, 42, 269–285.
Feng, X., Johansson, A. C., & Zhang, T. (2015). Mixing business with politics: Political participation by entrepreneurs in China. Journal of Banking & Finance, 59,
220–235.
Francis, B. B., Hasan, I., & Sun, X. (2009). Political connections and the process of going public: Evidence from China. Journal of International Money and Finance, 28,
696–719.
Frydman, R., Gray, C., Hessel, M., & Rapaczynski, A. (1999). Private ownership and corporate performance: Evidence from the transition economies. Quarterly Journal
of Economics, 114, 1153–1192.
Gormley, T. A., & Matsa, D. A. (2014). Common errors: How to (and not to) control for unobserved heterogeneity. Review of Financial Studies, 27, 617–661.
Guo, G. (2009). China's local political budget cycles. American Journal of Political Science, 53, 621–632.
Hsieh, C. T., & Song, M. F. (2016). Grasp the large, let go of the small: The transformation of the state sector in China. Brookings Papers on Economic Activity, 2015(1),
295–366.
Hsieh, C.-T., & Klenow, P. J. (2009). Misallocation and manufacturing TFP in China and India. Quarterly Journal of Economics, 124(4), 1403–1448.
Hu, X., & Liu, A. (2011). Political connections and access to IPO markets by privately-owned enterprises. 4, Caijing Luncong (Collected Essays on Finance and Economics)
62–67 (in Chinese).
Huyghebaert, N., & Wang, L. (2012). Expropriation of minority investors in Chinese listed firms: The role of internal and external corporate governance mechanisms.
Corporate Governance: An International Review, 20, 308–332.
Jain, B. A., & Kini, O. (1994). The post-issue operating performance of IPO firms. Journal of Finance, 49, 1699–1726.
Jia, R. (2017). Pollution for promotion. Mimeo.
Johansson, A. C., & Feng, X. (2016). The state advances, the private sector retreats? Firm effects of China's great stimulus program. Cambridge Journal of Economics (in
press).
Johansson, A. C., & Ljungwall, C. (2009). Spillover effects among the Greater China stock markets. World Development, 37, 839–851.
Kim, K. A., Kitsabunnarat, P., & Nofsinger, J. R. (2004). Ownership and operating performance in an emerging market: Evidence from Thai IPO firms. Journal of
Corporate Finance, 10, 355–381.
Kornai, J., Maskin, E., & Roland, G. (2003). Understanding the soft budget contraint. Journal of Economic Literature, 41(4), 1095–1136.
La Porta, R., & Lopez-de-Silanes, F. (1999). The benefits of privatization: Evidence from Mexico. Quarterly Journal of Economics, 114, 1193–1242.
La Porta, R., Lopez-de-Silanes, F., & Shleifer, A. (2002). Government ownership of banks. Journal of Finance, 57, 265–301.
Landry, P. F., Lu, X., & Duan, H. (2016). Does performance matter? Evaluating the institution of political selection along the Chinese Administrative ladder. Mimeo.
Landry, P. F. (2008). Decentralized authoritarianism in China: The Communist Party's control of local elites in the post-Mao era. New York: Cambridge University Press.
Li, G., & Zhou, H. (2015). Political connections and access to IPO markets in China. China Economic Review, 33, 76–93.
Li, H., & Zhou, L. (2005). Political turnover and economic performance: The incentive role of personnel control in China. Journal of Public Economics, 89, 1743–1762.
Lin, J. Y., Cai, F., & Li, Z. (1998). Competition, policy burdens, and state-owned enterprise reform. American Economic Review: Papers and Proceedings, 88, 422–427.
Lin, J. Y., & Tan, G. (1999). Policy burdens, accountability, and the soft budget constraints. American Economic Review: Papers and Proceedings, 89, 426–431.
Lin, L.-W., & Milhaupt, C. J. (2013). We are the (national) champions: Understanding the mechanisms of state capitalism in China. Stanford Law Review, 65(697), 735.
Lin, Y., & Li, Z. (2008). Policy burden, privatization and soft budget constraint. Journal of Comparative Economics, 36, 90–102.
Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (1997). Privatization in the United States. RAND Journal of Economics, 28, 447–471.
Maskin, E., Qian, Y., & Xu, C. (2000). Incentives, information, and organizational form. The Review of Economic Studies, 67(2), 359–378.
Megginson, W. L., Nash, R. C., & Randenborgh, M. (1994). The financial and operating performance of newly privatized firms: An international empirical analysis.
Journal of Finance, 49, 403–452.
Pagano, M., Panetta, F., & Zingales, L. (1998). Why do companies go public? An empirical analysis. Journal of Finance, 53, 27–64.
Piotroski, J. D., & Zhang, T. (2014). Politicians and the IPO decision: The impact of impending political promotions on IPO activity in China. Journal of Financial
Economics, 111, 111–136.
Ritter, J. R. (1991). The long-run performance of initial public offerings. Journal of Finance, 46, 3–27.
Ritter, J. R., & Welch, I. (2002). A review of IPO activity, pricing, and allocations. Journal of Finance, 57, 1795–1828.
Röller, L.-H., & Zhang, Z. (2005). Bundling of social and private goods and the soft budget constraint problem. Journal of Comparative Economics, 33(1), 47–58.
Schleifer, A., & Vishny, R. W. (1994). Politicians and firms. Quarterly Journal of Economics, 109(4), 995–1025.
Shleifer, A. (1998). State versus private ownership. Journal of Economic Perspectives, 12, 133–150.
Sun, L., & Liu, C. (2012). Capital Province, political objectives and the post-IPO policy burden. China Finance Review International, 2(2), 121–142.
Tsai, E. S. (2007). Capitalism without democracy: The private sector in contemporary China. Ithaca, NY: Cornell University Press.
Walter, C. E., & Howie, F. J. T. (2011). Red capitalism: The fragile foundation of China's extraordinary rise. Singapore: John Wiley & Sons.
Wang, X., Xu, L. C., & Zhu, T. (2004). State-owned enterprises going public: The case of China. The Economics of Transition, 12, 467–487.
Xia, L. J., & Fang, Y. Q. (2005). Government control, institutional environment and firm value: evidence from the Chinese securities market. Economic Research Journal,
5, 40–51 (in Chinese).
Xu, C. (2011). The fundamental institutions of China's reforms and development. Journal of Economic Literature, 49(4), 1076–1151.
Xu, L. C., Zhu, T., & Lin, Y. M. (2005). Politician control, agency problems and ownership reform. The Economics of Transition, 13, 1–24.
Zeng, Q., & Chen, X. (2006). State stockholder, excessive employment and labor cost. Economic Research Journal, 5, 74–86 (in Chinese).

281