Misallocation and Productivity: The Case - 麻省理工学院人工智能研究专业

Misallocation and Productivity: The Case
of Vietnamese Manufacturing
∗
DOAN THI THANH HA, KOZO KIYOTA, AND KENTA YAMANOUCHI

This paper attempts to measure the effect of resource misallocation on
aggregate manufacturing total factor productivity, focusing on Vietnamese
manufacturing ﬁrms during the period 2000–2009. One of the major ﬁndings of
this paper is that there would have been substantial improvement in aggregate
total factor productivity in Viet Nam in the absence of distortions. 这
results imply that potential productivity gains from removing distortions in
Vietnamese manufacturing are large. We also ﬁnd that smaller ﬁrms tend to face
advantageous distortions, while larger ﬁrms tend to face disadvantageous ones.
而且, the efﬁcient size distribution is more dispersed than the actual size
分配. These results suggest that Viet Nam’s policies may constrain its
largest and most efﬁcient producers, and coddle its smallest and least efﬁcient
那些.

关键词: misallocation, total factor productivity, Viet Nam
JEL codes: D22, F14, O47

我. 介绍

Differences in per capita income across economies result mainly from
differences in total factor productivity (TFP).1 所以, clarifying the underlying
causes of low productivity in developing economies is one of the central concerns
in various ﬁelds of economics such as development economics, international
经济学, and macroeconomics. Given the fact that production efﬁciency is
heterogeneous across ﬁrms, some recent studies on this issue argue that aggregate

∗Doan Thi Thanh Ha: Asian Development Bank Institute. 电子邮件: hato.doan@gmail.com; Kozo Kiyota (相应的
作者): Keio Economic Observatory, Keio University, 日本. 电子邮件: kiyota@sanken.keio.ac.jp; Kenta Yamanouchi:
Graduate School of Economics, Keio University, 日本. 电子邮件: ymdkntr40723331@yahoo.co.jp. The authors would
like to thank Iwan J. Azis, Flora Bellone, Minsoo Lee, Qing Liu, Atsushi Ohyama, Yoichi Sugita, Kenichi Ueda, 这
participants at the Asian Development Outlook–Asian Development Review Conference held in Seoul in November
2015, the managing editor, and an anonymous referee for helpful comments. Kozo Kiyota gratefully acknowledges
ﬁnancial support received from the Japan Society for the Promotion of Science Grant-in-Aid (26220503) 和
the Ministry of Education, Culture, Sports, 科学, and Technology-Supported Program for Strategic Research
Foundations at Private Universities. The usual disclaimer applies. ADB recognizes “China” as the People’s Republic
of China and “Vietnam” as Viet Nam.

1“Large differences in output per worker between rich and poor economies have been attributed, in no
small part, to differences in total factor productivity” (Hsieh and Klenow 2009, p. 1403); “[C]ross-economy income
differences mostly result from differences in total factor productivity” (Waugh 2010, p. 2095). McMillan and Rodrik
(2011) also argued for the importance of resource reallocation in enhancing productivity growth in developing
经济体.

Asian Development Review, 卷. 33, 不. 2, PP. 94–118

C(西德:3) 2016 Asian Development Bank
and Asian Development Bank Institute

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 95

TFP depends not only on the TFP of individual ﬁrms but also on the allocation of
resources across ﬁrms.2 In other words, low productivity in developing economies
can be attributed to the misallocation of resources across heterogeneous ﬁrms.

How do we measure the misallocation of resources? One way to answer this
question is to focus on distortions that reﬂect the difference between the actual
and efﬁcient outcomes. Such distortions are called “wedges” in the literature. 这
seminal work of Hsieh and Klenow (2009) estimates wedges from data on value
added and factor inputs for manufacturing establishments in the People’s Republic
of China (PRC), 印度, 和美国 (我们). They found that the distortions
were much larger in the PRC and India than in the US. Hsieh and Klenow (2009)
also found that the removal of distortions has a signiﬁcant effect on aggregate TFP
in the PRC and India. Following Hsieh and Klenow (2009), several studies have
provided a similar picture: large TFP gains could be expected from the removal of
distortions.3

This paper extends the analysis of Hsieh and Klenow (2009) to Vietnamese

manufacturing between 2000 和 2009 and asks the following questions:

(我) To what extent are resources misallocated in Viet Nam?

(二) How large would the productivity gains have been in the absence of

distortions?

(三、) Are the distortions related to ﬁrm size?

(四号) What would the distribution of ﬁrm size have been in the absence of

distortions?

Answering these questions has important implications for potential growth
because reallocation would lead to productivity gains that can accelerate potential
growth through improved interﬁrm resource allocation.

Our study is closely related to Bach (2014), who also examined resource
misallocation in Viet Nam using ﬁrm-level data. His study addressed the ﬁrst two
questions above but did not compare resource misallocation in Viet Nam with
misallocation in other Asian economies. Nor did his study address the last two
问题. From a policy perspective, the last two questions are important because
many economies give preferential treatment to small and medium-sized enterprises
(中小企业). 的确, size-dependent policies, which limit the size of ﬁrms, could be an

2See Restuccia and Rogerson (2013) and Hopenhayn (2014) for a survey.
3看, 例如, Camacho and Conover (2010) for the case of Colombia; Busso, Madrigal, and Pages–Serra
(2012) for Latin America; Bellone and Mallen–Pisano (2013) for France; Hosono and Takizawa (2013) for Japan; 的
Vries (2014) for Brazil; Dheera–Aumpon (2014) for Thailand; 巴赫 (2014) for Viet Nam; and Calligaris (2015) 为了
意大利.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

96 ASIAN DEVELOPMENT REVIEW

important source of misallocation (Restuccia and Rogerson 2013). In answering the
four questions, this paper goes one step further by providing a deeper understanding
of the potential productivity gains from removing distortions in Viet Nam.4

The rest of this paper is organized as follows. In section II, we describe
the methodology of Hsieh and Klenow (2009). Section III describes the Vietnamese
ﬁrm-level data used in our study. Section IV presents the results. Concluding remarks
and policy implications are presented in section V.

二. Measurement of Misallocation

Hsieh and Klenow (2009) formulated an analytical framework to estimate
misallocation. Although some studies such as Bartelsman, Haltiwanger, 和
Scarpetta (2013) developed an alternative framework, this paper employs Hsieh and
Klenow’s framework for the following reasons. 第一的, their framework is tractable
in the sense that it is simple and its data requirements are minimal. This provides a
signiﬁcant advantage in estimating misallocation in Viet Nam because of the limited
data availability, as we will discuss in the next section. 第二, the framework allows
us to decompose the source of misallocation into distortions in output markets and
those in capital markets. Such decompositions are useful if the distortions come
from different sources. The Hsieh and Klenow (2009) methodology is summarized
以下.

Assume that a representative ﬁrm produces a single ﬁnal good, 是 , 在一个
perfectly competitive ﬁnal goods market. The ﬁrm produces Y , using the output
Ys of S manufacturing industries, with the following Cobb–Douglas production
技术:

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

Y =

S(西德:2)

s=1

Y θs
s

, 在哪里

S(西德:3)

s=1

θs = 1

(1)

and θs is the output share of each industry s.

Each industry produces output, Ys, using Ms differentiated goods produced by
individual ﬁrm i with a constant elasticity of substitution technology (s = 1, . . . , S).
Output in industry s is then given by:5

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Ys =

(西德:4)

Ms(西德:3)

我=1

(西德:5) σ
σ −1

σ −1
σ

是

σ > 1

(2)

4Another important difference between Bach (2014) and our study is that his study did not control for the

skill differences of workers across ﬁrms in measuring quantity-based TFP and revenue-based TFP.

5We suppress the time subscript to avoid heavy notation, although we utilize ﬁrm-level panel data in the

empirical analysis.

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 97

where σ is the elasticity of substitution between varieties and Ysi is the output of the
differentiated good produced by ﬁrm i in industry s, using capital and labor, 基于
on the following Cobb–Douglas technology:

Ysi = Asi K

si L 1−αs
作为

(3)

where Asi , Ksi , and L si denote the productivity, 首都, and labor of ﬁrm i in
industry s, 分别; and αs represents the capital share, which is different across
industries but the same across ﬁrms within an industry.

To assess the extent of misallocation, Hsieh and Klenow (2009) followed
促进, Haltiwanger, and Syverson (2008) in making a distinction between physical
生产率, denoted by TFPQ, and revenue productivity, denoted by TFPR:

TFPQsi

(西德:5)= Asi =

Ysi
si L 1−αs
作为

和

TFPRsi

(西德:5)= Psi Asi = Psi Ysi
si L 1−αs
作为

(4)

(5)

分别, where Psi represents the ﬁrm-speciﬁc output price.

In addition to ﬁrm heterogeneity in terms of productivity (看, 例如,
Melitz 2003), ﬁrms potentially face different output and capital distortions. 更多的
具体来说, Hsieh and Klenow (2009) incorporated two types of ﬁrm-level wedges
into this framework. One raises the marginal product of capital and labor by the
same proportion, which is denoted by τY si . The other increases the marginal product
of capital relative to labor, which is denoted by τK si . These wedges are given from
the ﬁrm’s viewpoint and we do not make any assumptions about what generates
them.6

An example of such distortions is subsidized credit. If two ﬁrms have identical
technologies but one of the ﬁrms can borrow from the ﬁnancial market at a lower
interest rate (and the other ﬁrm can borrow at a higher interest rate), the marginal
product of capital of the ﬁrm that can access the subsidized credit will be lower than
that of the other ﬁrm. This results in the misallocation of capital because one ﬁrm
enjoys a lower interest rate even though the two ﬁrms have the same technologies.

6Distortions can be generated by various factors such as trade policies and credit market imperfections. In our
companion paper (Ha and Kiyota 2015), we examined the determinants of distortions in Vietnamese manufacturing.
Le´on–Ledesma and Christopoulos (2016) examined the effects of access to ﬁnance obstacles on misallocation. 使用
ﬁrm-level data covering 45 经济体, they found that access to ﬁnance obstacles and private credit increase the
dispersion of distortions. 然而, they also found that the ﬁnancial variables explain a small part of the dispersion
of factor market and size distributions.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

98 ASIAN DEVELOPMENT REVIEW

换句话说, in the framework of Hsieh and Klenow (2009), the differences in
factor prices mean the existence of distortions.

With these wedges, the expected proﬁts of the ﬁrm are written as follows:7

πsi = (1 − τY si ) Psi Ysi − wL si − (1 + τK si ) R Ksi

(6)

where w and R denote the common wages and rental costs facing all ﬁrms,
分别. Firms maximize their proﬁts under the following constraint:

(西德:7)σ

(西德:6)

Ps
Psi

Ysi = Ys

在哪里

Ps ≡

(西德:5) 1
1−σ

(西德:4)

Ms(西德:3)

我=1

P 1−σ
si

(7)

(8)

In the presence of distortions, ﬁrms will produce a different quantity compared with
what they would produce without these wedges (the efﬁcient case).

Solving the proﬁt maximization problem under a monopolistic competition

framework and the equilibrium allocation of resources across industries, 我们有:

Psi =

1 − τY si =

1 + τK si =

σ
σ − 1
σ
σ − 1
作为
1 − αs

(西德:7)作为

(西德:6)

右
作为

w
1 − αs

(西德:7)

1−αs

A−1
si

(1 + τK si )作为
1 − τY si

, 和

wL si
(1 − αs) Psi Ysi
wL si
R Ksi

From equation (9), 我们有:

TFPRsi = ξs

(1 + τK si )作为
1 − τY si

在哪里

ξs =

(西德:7)作为

(西德:6)

σ
σ − 1

(西德:6)

右
作为

w
1 − αs

(西德:7)

1−αs

(9)

(10)

(11)

(12)

(13)

7Distortions to output and to capital relative to labor are an observationally equivalent characterization of
distortions to the absolute levels of capital and labor. For more details, see Hsieh and Klenow (2009, Appendix III).

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 99

Noting that ξs is different across industries but constant within an industry, 方程
(12) implies:

TFPRsi ∝

(1 + τK si )作为
1 − τY si

(14)

This equation means that the large deviation of ﬁrm TFPR from ξs is a sign that the
ﬁrm faces large distortions.

If we denote industry TFP as TFPs and deﬁne industry TFP as a weighted

geometric average of ﬁrm i’s TFPQsi , 我们有

⎡

Ms(西德:3)

⎣

(西德:5)=

我=1

TFPs

(西德:4)

TFPQsi

TFPRs
TFPRsi

(西德:5)σ −1

1
σ −1

⎤

⎦

(15)

where TFPRs is the geometric average of the average marginal revenue product of
labor and capital in industry s:

(西德:12)

TFPRs

(西德:5)=

σ
σ − 1

(西德:13)

Ms
我=1

作为

右
1−τY si
1+τK si

Psi Ysi
Ps Ys

(西德:14)作为

(西德:12)

(1 − αs)

(西德:13)

(西德:14)

1−αs

Ms
我=1 (1 − τY si ) Psi Ysi
Ps Ys

(16)

There are two points of clariﬁcation regarding equation (15). 第一的, 越高
dispersion in TFPR, the lower the industry TFP will be. Hsieh and Klenow (2013)
showed that when TFPQ and TFPR are jointly log-normally distributed and when
there is only variation in log (1 − τY si ), aggregate TFP can be expressed as follows:8

logTFPs = 1

σ − 1

(西德:15)
log Ms + log E

(西德:16)

TFPQσ −1

(西德:17)(西德:18)

var (log TFPRsi )

(17)

This equation suggests that industry TFP will decline if the elasticity of substitution
σ and/or TFPR dispersion increase.

第二, TFPR will be equalized across ﬁrms within industry s if τK si and τY si
are equalized. 例如, from equation (12), TFPRsi = ξs∀i if τK si = τY si = 0.
This implies that TFPRsi = ξs = TFPRs∀i.9 Denoting industry TFP without any

8A similar property is obtained even when there is variation in log (1 + τK si ), although the equation becomes

more complicated. For more details, see Hsieh and Klenow (2013).

9Note that even when TFPR is equalized across ﬁrms, TFPQ can be different across ﬁrms because more
productive ﬁrms charge lower prices (参见方程 [9]); 那是, if Asi > As j and Psi < Ps j , Psi Asi could be equal to Ps j As j for i (cid:7)= j. l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a d e v / a r t i c e - p d l f / / / / / 3 3 2 9 4 1 6 4 3 6 6 2 a d e v _ a _ 0 0 0 7 4 p d . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 100 ASIAN DEVELOPMENT REVIEW distortions as TFPQs from equation (15), we can obtain TFPQs (cid:5)= ¯As = (cid:5) 1 σ −1 (cid:4) Ms(cid:3) i=1 Aσ −1 si (18) which is called “efﬁcient” industry TFP. In order to obtain “efﬁcient” TFP, one needs information on ﬁrm-level TFPQ (Asi ). One problem is the limited availability of ﬁrm-level price data, Psi , which are not available for many economies, including Viet Nam.10 Hsieh and Klenow (2009) rewrote equation (4) as TFPQsi = Asi = κs σ σ −1 (Psi Ysi ) si L 1−αs αs K si , where κs = w1−αs σ −1 (PsYs)− 1 Ps (19) Noting that κs is a scaling constant by industry and does not affect the relative differences between ﬁrms within industry s, it can be normalized to unity (κs = 1). This manipulation enables us to estimate TFPQ without ﬁrm-level price data. > TFPRsi if κs = 1 and Psi Ysi ≥ 1.
Note that from equations (5) 和 (19), TFPQsi
所以, in the Hsieh and Klenow (2009) framework the dispersion of TFPQ tends
to be larger than that of TFPR.

三、. 数据

来源

This paper utilizes ﬁrm-level data from the Annual Survey of Enterprises
collected by the General Statistics Ofﬁce of Viet Nam.11 The survey was conducted
for the ﬁrst time in 2000 and then annually thereafter to provide researchers and
policy makers with comprehensive information on Vietnamese ﬁrms. These data
cover registered ﬁrms operating in all sectors, including agriculture, industry and
建造, and services.

The survey covers all state-owned enterprises and foreign-owned ﬁrms
without any ﬁrm size threshold. 然而, for domestic private ﬁrms, those with
少于 10 workers are chosen by random sampling.12 Household business

10There are some economies for which ﬁrm-level (or plant-level) price data are available. 例如, Eslava

等人. (2004) utilized plant-level price data for Colombia to estimate plant-level TFPQ.

11We use the same data as Ha and Kiyota (2014); this section is based on section III of their study. Note also
that the use of ﬁrm-level data is more consistent with the theory than the use of plant-level data. 这是因为, 作为
Nishimura, Nakajima, and Kiyota (2005) point out, resource allocation within a ﬁrm is determined by managerial
决定. 而且, research and development and headquarters activities are typically classiﬁed as service activities,
which are not covered in the manufacturing survey.

12This threshold was used in surveys before 2010. 从 2010, different regions set different ﬁrm size

thresholds.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 101

activities are not covered in this survey.13 The survey information includes type of
所有权, assets and liabilities, number of employees, sales, capital stock, 行业
that the ﬁrm belongs to, and obligations to the government (例如, 税收) from January
to December of that year.

The data have some disadvantages. Some of the input data, such as materials,
are not available for all years. Information on working hours and capital utilization
rates is also unavailable. Firms’ year of establishment and export status are not
available every year. This paper uses ﬁrms with information on inputs, outputs, 和
cost shares. Reentry ﬁrms, which are those that disappeared from the data and then
reappeared later, are omitted from our analysis. Some ﬁrms changed industry and/or
ownership during the sample period.14 We drop ﬁrms with fewer than 10 雇员,
regardless of their ownership, to avoid the effects of the random sampling.

乙.

Variables and Parameters

The main variables that we use are the two-digit Viet Nam Standard Industry
Classiﬁcation (VSIC) industry code, ownership type, value added, 就业,
total labor costs, and capital stock. Following Hsieh and Klenow (2009), 我们用
wage bills instead of the number of workers to capture the potential differences in
employee quality.15 Capital stock is measured as total ﬁxed assets recorded at the
end of each year. Both wage bills and capital stock are deﬂated by the manufacturing
国内生产总值 (GDP) deﬂator.16

To compute dispersion, we follow other research in setting the key parameters
σ and R as follows. We assume that the elasticity of substitution σ equals 3 and R is
10%, comprising a 5% depreciation rate and a 5% interest rate. We also follow Hsieh
and Klenow (2009) to set αs equal to 1 minus the labor share in the corresponding
industry in the US. Under Hsieh and Klenow’s framework, the output elasticities of
capital and labor (αs and 1 − αs) do not embed distortions. Given the assumption
that the US economy is less distorted than the Vietnamese economy, the use of US
shares can be justiﬁed.

The US labor share is obtained from the NBER–CES Manufacturing Industry
Database, which is a joint product of the National Bureau of Economic Research
and the US Census Bureau’s Center for Economic Studies.17 Industry classiﬁcations

13The survey covered 62.2% of total employment in manufacturing in 2009. The data on total employment

in manufacturing were obtained from the General Statistics Ofﬁce online database on population and employment.

14If a ﬁrm has switched industries, the industry to which the ﬁrm belonged for the majority of the surveyed
years is regarded as the ﬁrm’s industry. If a ﬁrm belonged to more than one industry for equal amounts of time, 我们
assign the industry code of the industry that the ﬁrm belonged to most recently.

15The use of wage bills as a measure of labor input implies that w = 1 (Camacho and Conover 2010, p. 10).
16As Aw, 陈, and Roberts (2001) pointed out, it is preferable to utilize the investment goods price deﬂator
rather than the manufacturing GDP deﬂator to obtain the real capital stock. 然而, as Ha and Kiyota (2014)
discussed, the investment goods price deﬂator is not available for our data set.

17Data can be downloaded from the National Bureau of Economic Research’s website at http://www.nber.org

/nberces/

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

102 ASIAN DEVELOPMENT REVIEW

are based on the North American Industry Classiﬁcation System (NAICS) version
1997. Based on the data, we ﬁrst match the NAICS code with the four-digit VSIC
code using concordance tables between NAICS, International Standard Industry
Classiﬁcation revision 3, and VSIC. We then aggregate total payroll and total value
added by two-digit VSIC sectors. To compute the labor share, we take the ratio of total
payroll over total value added by sector. Because total payroll in the database does
not include fringe beneﬁts and employer’s contribution to social security, this labor
share only reﬂects two-thirds of the aggregate labor share in the whole manufacturing
sector. 所以, we follow Hsieh and Klenow (2009) to inﬂate the labor shares by
1.5 to obtain US labor elasticities.

As ﬁrms’ output prices are not available, we have obtained TFPQ by
raising nominal output to the power of σ/(σ − 1), assuming that normal demand
relationships hold. If a ﬁrm’s real output is high, one would expect its price
to be low so that consumers demand more output. Following Ziebarth (2013),
the dispersion of TFP is deﬁned as the deviation of the log of TFP from its
industry mean: 日志(TFPRsi /TFPRs) and log(TFPQsi M
/TFPQs), where TFPRs
and TFPQs are from equations (16) 和 (18), respectively.18 We trim 2% 的
ﬁrm productivity and distortions by removing values below the ﬁrst percentile
and above the 99th percentile from the distribution of log(TFPRsi /TFPRs) 和
/TFPQs). 然后, we recalculate TFPRs, TFPQs, and TFPs. 为了
日志(TFPQsi M
robustness checks, section V examines whether the results are sensitive to the values
of σ , 作为, and the threshold level of trimming.

1
σ −1
s

IV. 结果

To what extent are resources misallocated in Viet Nam?

This section addresses the ﬁrst question of the paper: To what extent are
resources misallocated in Viet Nam? To answer this question, we compare the
dispersions of TFP in Viet Nam with those in the PRC, 印度, 日本, Thailand, 和
the US. The dispersions of TFPR are reported in Table 1, while those of TFPQ are
reported in Table 2. Both tables present standard deviations, differences between the
90th and 10th percentiles, differences between the 75th and 25th percentiles, 和
average per capita GDP during the sample period.19 Figures for the PRC, 印度, 和
the US are from Hsieh and Klenow (2009); for Japan, from Hosono and Takizawa
(2013); and for Thailand, from Dheera–Aumpon (2014).

These tables indicate that the standard deviation of TFPR for Viet Nam is
0.79, which is comparable to the standard deviations for the PRC (0.68), 印度 (0.68),

18Some of the effects of the changes in prices are controlled for by taking the ratio.
19Noting that both TFPR and TFPQ are divided by their industry means, these statistics can be interpreted as

the coefﬁcients of variation.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 103

桌子 1. Dispersion of Revenue-Based Total Factor Productivity

Viet Nam Thailand
2000–2009

2006

People’s
Republic of
中国
1998–2005年

印度
1987–1994

日本
1981–2008

团结的
状态
1977–1997

0.68
0.80
1.66
400

0.79
0.97
2.00
685

0.68
0.89
1.72
1,304

0.85
1.04
2.09
2,813

0.55
0.70
1.40
31,101

0.45
0.47
1.08
30,533

标清
75–25
90–10
GDP per capita
GDP = gross domestic product, SD = standard deviation, TFPR = revenue-based total factor productivity.
Notes: Data for Thailand are from Dheera–Aumpon (2014, 桌子 3). Data for the People’s Republic of China are
arithmetic averages from Hsieh and Klenow (2009, 桌子 2). Data for Japan are from Hosono and Takizawa (2013).
TFPR is calculated from equation (5) and then scaled by the geometric mean of TFPR across all ﬁrms in an industry
s. Industries are weighted by value-added shares. GDP per capita is the annual average over each sample period in
持续的 2005 US dollars.
来源: Hsieh, C.-T., 和P. J. Klenow. 2009. Misallocation and Manufacturing TFP in China and India. 季刊
经济学杂志 124 (4): 1403–48; Hosono, K., 和M. Takizawa. 2013. Misallocation and the Dynamics
of Establishment. Financial Review 112 (1): 180–209 (in Japanese); Dheera–Aumpon, S. 2014. Misallocation
and Manufacturing TFP in Thailand. Asia-Paciﬁc Economic Literature 28 (2): 63–76; and authors’ calculations
based on Government of Viet Nam, General Statistics Ofﬁce. Annual Survey of Enterprises. https://www.gso.gov.vn
/default_en.aspx?tabid=479&idmid=5; per capita GDP data obtained from World Bank. 2014. World Development
指标. 华盛顿, 直流.

桌子 2. Dispersion of Quantity-Based Total Factor Productivity

People’s

Viet Nam Thailand Republic of China
2000–2009

1998–2005年

2006

印度
1987–1994

日本
1981–2008

美国
1977–1997

1.59
2.18
4.12

1.42
2.01
3.70

0.98
1.27
2.48

1.00
1.34
2.57

标清
1.19
75–25
1.56
90–10
3.03
SD = standard deviation, TFPQ = quantity-based total factor productivity.
Notes: Data for Thailand are from Dheera–Aumpon (2014, 桌子 2). Data for the People’s Republic of China, 印度,
and the United States are arithmetic averages from Hsieh and Klenow (2009, 桌子 1). Data for Japan are from
Hosono and Takizawa (2013, 桌子 1). TFPQ is calculated from equation (19) and then scaled by the geometric mean
of TFPQ across all ﬁrms in an industry s. Industries are weighted by value-added shares.
来源: Hsieh, C.-T., 和P. J. Klenow. 2009. Misallocation and Manufacturing TFP in China and India. 季刊
经济学杂志 124 (4): 1403–48; Hosono, K., 和M. Takizawa. 2013. Misallocation and the Dynamics
of Establishment. Financial Review 112 (1): 180–209 (in Japanese); Dheera–Aumpon, S. 2014. Misallocation
and Manufacturing TFP in Thailand. Asia-Paciﬁc Economic Literature 28 (2): 63–76; and authors’ calculations
based on Government of Viet Nam, General Statistics Ofﬁce. Annual Survey of Enterprises. https://www.gso.gov.vn
/default_en.aspx?tabid=479&idmid=5

0.83
1.16
2.15

and Thailand (0.85), and is larger than the standard deviations for Japan (0.55) 和
the US (0.45). Similar patterns were also conﬁrmed for the differences between the
75th and 25th percentiles, and between the 90th and 10th percentiles.20 Although
more careful examination is needed in the form of a direct comparison, 结果

20The difference between the 75th and 25th percentile ﬁrms is 0.97, which corresponds to a TFP ratio of
e0.97 = 2.63. 相似地, the difference between the 90th and 10th percentile ﬁrms is 2, which corresponds to a TFP
ratio of e2.00 = 7.39. These ﬁgures are much larger than those for the US. For more details, see Syverson (2011).

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

104 ASIAN DEVELOPMENT REVIEW

suggest that distortions in developing economies, including Viet Nam, tend to be
large relative to those in developed economies.

乙.

How large would the productivity gains be without distortions?

This section addresses the second question of this paper: How large would the
productivity gains have been in the absence of distortions? To answer this question,
we estimate TFP gains when the marginal products of labor and capital are equalized
across ﬁrms within each industry. For each industry, the gains are computed as the
ratio of actual TFP obtained from equation (15) to the “efﬁcient” TFP obtained
from equation (18). We then aggregate the gains across industries using industry
value-added shares as the weights. 尤其, we compute

(西德:5)=

是
Y ∗

(西德:6)

S(西德:2)

s=1

S(西德:2)

s=1

⎧
⎪⎨

⎪⎩

⎡

(西德:6)

S(西德:2)

(西德:7)θs

s=1

TFPs
TFPQs

(西德:7)θs

Ys
Y ∗
s

⎡

⎣

(西德:4)

Ms(西德:3)

TFPQsi

我=1

1
TFPQs

TFPRs
TFPRsi

(西德:5)σ −1

1
σ −1

⎤

⎦

θs

⎫
⎪⎬

⎪⎭

S(西德:2)

Ms(西德:3)

⎣

s=1

我=1

(西德:4)

Asi
¯As

TFPRs
TFPRsi

(西德:5)σ −1

θs
σ −1

⎤

⎦

(20)

where Y ∗ is the “efﬁcient” output that corresponds to the “efﬁcient” TFP and θs is
= TFPs/
the value-added share of industry s (
TFPQs) is obtained when Ks and L s are given. As the total amount of inputs is ﬁxed,
the output gains come solely from the reallocation of resources in the absence of
distortions.

θs = 1). The ﬁrst equality (Ys/Y ∗
s

(西德:13)
s

桌子 3 presents the TFP gains from equalizing TFPR across ﬁrms within
each industry. The gains are measured relative to the TFP gains in the US in 1997.21
To report TFP percentage gains in Viet Nam relative to those in the US, we take
the ratio of Y ∗/Y to the US equivalent in 1997, subtract 1, and multiply by 100.
If Viet Nam hypothetically moves to “US efﬁciency,” substantial gains (30.7%) 是
预期的. The gains are smaller than those for the PRC (39.2%), 印度 (46.9%), 和
Thailand (73.4%), but larger than those for Japan (3%).

One may be concerned that the dispersion of TFPR is larger (桌子 1), 尽管
the gains are smaller in Viet Nam than in either the PRC or India (桌子 3). Noting
that the gains are computed from the inverse of equation (20), (Y ∗/Y − 1) × 100),

21Hsieh and Klenow (2009) called this comparison a conservative analysis because the US’ gains are largest

在 1997.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 105

桌子 3. Total Factor Productivity Gains from Equalizing Revenue-Based Total
Factor Productivity Relative to 1997 Gains in the United States

Viet Nam
2000–2009

Thailand
2006

People’s
Republic of China
1998–2005年

印度
1987–1994

日本
1981–2008

30.7

73.4

39.3

46.9

3.0

Notes: The data for Thailand are calculated from Dheera–Aumpon (2014, 桌子 4). The data for the
People’s Republic of China, 印度, and the United States are arithmetic averages from Hsieh and Klenow
(2009, 桌子 6). The data for Japan are calculated from Hosono and Takizawa (2013, 桌子 2).
来源: Hsieh, C.-T., 和P. J. Klenow. 2009. Misallocation and Manufacturing TFP in China and
印度. Quarterly Journal of Economics 124 (4): 1403–48; Hosono, K., 和M. Takizawa. 2013.
Misallocation and the Dynamics of Establishment. Financial Review 112 (1): 180–209 (in Japanese);
Dheera–Aumpon, S. 2014. Misallocation and Manufacturing TFP in Thailand. Asia-Paciﬁc Economic
Literature 28 (2): 63–76; and authors’ calculations based on Government of Viet Nam, General Statistics
Ofﬁce. Annual Survey of Enterprises. https://www.gso.gov.vn/default_en.aspx?tabid=479&idmid=5

Y ∗/Y will be small if Asi / ¯As and/or TFPRs/TFPRsi become large. The results
suggest that, 一般, Asi / ¯As is larger in Viet Nam than in either the PRC or
印度. 相似地, we ﬁnd large TFP gains for Thailand, which is possibly attributed
to a small Asi / ¯As for Thailand.22 Although these are hypothetical exercises and thus
should not be taken literally, the results suggest that substantial productivity gains
are expected in Viet Nam by the kind of reallocation considered here.

Are the distortions related to ﬁrm size?

This section examines whether the distortions are related to ﬁrm size. 这
question has important policy implications because, 例如, many economies
give preferential treatment to SMEs. If SMEs tend to face larger disadvantageous
distortions, preferential treatment to SMEs can be justiﬁed. Following Hsieh and
Klenow (2009) and Ziebarth (2013), we examine the relationship between ﬁrm size
and TFPR.

数字 1 presents the relationship between ﬁrm size percentile as measured
by value added and scaled TFPR relative to a given industry. 数字 1 indicates
that TFPR is increasing as ﬁrm size increases. Noting that TFPR is proportional
to the distortions (方程 14), this result implies that smaller ﬁrms tend to face
advantageous distortions, while larger ﬁrms tend to face disadvantageous ones. 这
result is similar to that found for India (Hsieh and Klenow 2009, 数字 6) 并为
the US in the 19th century (Ziebarth 2013, 数字 3).

有趣的是, this correlation with ﬁrm size is different for the distortions
in output and the distortions in capital markets. 数字 2 presents the relationship
between the distortions in output markets and ﬁrm size (in terms of value added).

22数字 1 in Dheera–Aumpon (2014) suggests that the distribution of TFPQ in Thailand moves to the left
and its mean takes a negative value. Although it is not clear why the distribution moves to the left, this may be a
reason why large TFP gains are expected in Thailand.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

106 ASIAN DEVELOPMENT REVIEW

数字 1. Revenue-Based Total Factor Productivity and Size

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

TFPR = revenue-based total factor productivity.
笔记: This ﬁgure presents the relationship between scaled TFPR relative to a given industry and size percentile as
measured by value added.
来源: Authors’ calculations based on Government of Viet Nam, General Statistics Ofﬁce. Annual Survey of
Enterprises. https://www.gso.gov.vn/default_en.aspx?tabid=479&idmid=5

数字 2 indicates that the distortions in output markets decrease as ﬁrm size
增加. Noting that the distortions in output markets are measured by 1 − τY ,
this result is similar to that in TFPR: smaller ﬁrms tend to face advantageous
distortions, while larger ﬁrms tend to face disadvantageous ones.

数字 3 presents the relationship between the distortions in capital markets
and ﬁrm size. In contrast to the distortions in output markets, 数字 3 节目
an inverse U-shaped relationship. Noting that the distortions in capital markets
are measured by 1 + τK , this result suggests that both small and large ﬁrms
tend to face advantageous distortions. 相比之下, middle-sized ﬁrms tend to
face disadvantageous distortions. This pattern is different from those of TFPR
and distortions in output markets. This may be because small ﬁrms are treated
preferentially, while large ﬁrms can diversify their capital procurement.

It is also interesting to note that the result for TFPR mainly reﬂects that
of distortions in output markets. This result implies that the distortions in output
markets have stronger effects on TFPR than those in capital markets. This result
is consistent with the ﬁndings of Midrigan and Xu (2014), who showed that
ﬁnancial frictions, measured by borrowing constraints, had relatively small impacts
on productivity.

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 107

数字 2. Distortions in Output Markets and Size

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

笔记: This ﬁgure presents the relationship between scaled 1 − τY relative to a given industry and size percentile as
measured by value added.
来源: Authors’ calculations based on Government of Viet Nam, General Statistics Ofﬁce. Annual Survey of
Enterprises. https://www.gso.gov.vn/default_en.aspx?tabid=479&idmid=5

One may be concerned that our measurement of ﬁrm size, following Hsieh and
Klenow (2009), is based on value added rather than employment. In many economies,
SMEs are deﬁned by the number of employees rather than by the size of their value
added. To address this concern, we examine the relationship between distortions and
ﬁrm size as measured by employment. The results are presented in Figures 4, 5, 和
6. The results are different from—but qualitatively similar to—those when ﬁrm size
is measured by value added: as ﬁrm size (in terms of employment) 增加, TFPR
is increasing, the distortions in output markets are decreasing, and the distortions
in capital markets show an inverse U-shaped relationship except for the top quintile
of ﬁrms. Noting that the results for TFPR mainly reﬂect the distortions in output
市场, we can conclude that our main messages remain unchanged even when
ﬁrm size is measured by employment.

D. What would the distribution of ﬁrm size have been in the absence

of distortions?

The model also has an implication for the distribution of ﬁrm size. 方程

(7) is rewritten as

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

108 ASIAN DEVELOPMENT REVIEW

数字 3. Distortions in Capital Markets and Size

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

笔记: This ﬁgure presents the relationship between scaled 1 + τK relative to a given industry and size percentile as
measured by value added.
来源: Authors’ calculations based on Government of Viet Nam, General Statistics Ofﬁce. Annual Survey of
Enterprises. https://www.gso.gov.vn/default_en.aspx?tabid=479&idmid=5

Psi Ysi = Y

σ −1
σ

1
σ
si PsY
s

From equations (7) 和 (9), 我们有

(西德:12)

σ − 1
σ

(西德:26) 作为
右

(西德:6)

(西德:27)作为

1 − αs
w

Ysi =

(西德:14)σ

(西德:7)

1−αs

(西德:28)

P σ
s Ys

(西德:29)σ

Asi (1 − τY si )
(1 + τK si )作为

Similar to equation (14), 从方程 (21) 和 (22), 我们有

Psi Ysi ∝

(西德:28)

Asi (1 − τY si )
(1 + τK si )作为

(西德:29)σ −1

(21)

(22)

(23)

方程 (23) suggests that without distortions, 更多的 (较少的) productive ﬁrms tend
to be larger (较小). When Asi and 1 − τY si are correlated negatively, 更多的
productive ﬁrms tend to be smaller than the efﬁcient size. 相似地, if Asi and
1 + τK si are correlated positively, less productive ﬁrms tend to be larger than the

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 109

数字 4. Revenue-Based Total Factor Productivity and Employment Size

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

TFPR = revenue-based total factor productivity.
笔记: This ﬁgure presents the relationship between scaled TFPR relative to a given industry and size percentile as
measured by employment.
来源: Authors’ calculations based on Government of Viet Nam, General Statistics Ofﬁce. Annual Survey of
Enterprises. https://www.gso.gov.vn/default_en.aspx?tabid=479&idmid=5

efﬁcient size. Both cases result in smaller size dispersion. This implies that when
distortions are large, the efﬁcient size distribution is more dispersed than the actual
size distribution.

To examine this implication, we compare the actual ﬁrm size distribution
with the efﬁcient ﬁrm size distribution. The size is measured as the value added of
the ﬁrms, following Hsieh and Klenow (2009). Let P ∗
si be the efﬁcient ﬁrm size.
The efﬁcient sizes relative to actual sizes are
(西德:29)σ −1
(西德:7)σ −1

si Y ∗

(西德:6)

(西德:28)

si Y ∗
P ∗
si
Psi Ysi

= Y ∗
是

Ys
Y ∗
s

(1 + τK si )作为
1 − τY si

(24)

where the efﬁcient ﬁrm size is obtained when τK si and τY si are equalized within
industry s. Both Y ∗/Y and Ys/Y ∗
s are obtained from equation (20).23 We compute
the actual and efﬁcient sizes from this equation by year, and then take averages over
the period.

23For the derivation of equation (24), see the Appendix.

110 ASIAN DEVELOPMENT REVIEW

数字 5. Distortions in Output Markets and Employment Size

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

笔记: This ﬁgure presents the relationship between scaled 1 + τY relative to a given industry and size percentile as
measured by employment.
来源: Authors’ calculations based on Government of Viet Nam, General Statistics Ofﬁce. Annual Survey of
Enterprises. https://www.gso.gov.vn/default_en.aspx?tabid=479&idmid=5

桌子 4 和图 7 present the results. 表中 4, the rows are the actual
ﬁrm size quartiles with equal numbers of ﬁrms. The columns are the bins of efﬁcient
ﬁrm size relative to actual ﬁrm size. We classify ﬁrms into four bins. 例如,
0%–50% means that the ﬁrm size would be less than half of the actual ﬁrm size if all
distortions are removed. 相似地, 200+% means that the ﬁrm size would be more
than double without distortions. The entries are the shares of ﬁrms (平均超过
the period). The rows sum to 25%; the rows and columns together sum to 100%.

Examining Table 4, we highlight two results. 第一的, although average output
rises substantially (as we conﬁrmed in section IV), many ﬁrms of all sizes would
shrink without distortions. 第二, the largest quartile indicates the largest expansion
among all ﬁrm sizes (8.7%). This result means that large ﬁrms are less likely to shrink
and more likely to expand. This ﬁnding is also conﬁrmed by Figure 7.

As the model suggests, the efﬁcient size distribution is more dispersed than
the actual size distribution. This result is consistent with the ﬁnding of the previous
部分. Like the case of India (Banerjee and Duﬂo 2005, p. 507), Viet Nam’s policies
may constrain its largest and most efﬁcient producers and coddle its smallest and
least efﬁcient ones. 的确, Vietnamese SMEs are supported by various policies
such as government-supported ﬁnancing (特兰, Le, and Nguyen 2008, PP. 347–59).

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 111

数字 6. Distortions in Capital Markets and Employment Size

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

笔记: This ﬁgure presents the relationship between scaled 1 + τK relative to a given industry and size percentile as
measured by employment.
来源: Authors’ calculations based on Government of Viet Nam, General Statistics Ofﬁce. Annual Survey of
Enterprises. https://www.gso.gov.vn/default_en.aspx?tabid=479&idmid=5

桌子 4. Actual Size versus Efﬁcient Size

Efﬁcient ﬁrm size relative to actual ﬁrm size

2000–2009 (average)

0%–50% 50%–100% 100%–200% 200%+ 全部的

Actual ﬁrm size
Top quartile
Second quartile
Third quartile
Bottom quartile

全部的

5.1
8.0
9.1
13.7

36.0

5.5
5.6
6.3
5.1

22.4

5.7
4.6
4.4
3.0

17.6

8.7
6.8
5.2
3.1

25.0
25.0
25.0
25.0

23.9

100.0

Notes: The rows are the actual ﬁrm size quartiles with equal numbers of ﬁrms. The columns are the
bins of efﬁcient ﬁrm size relative to actual ﬁrm size. We classify ﬁrms into four bins by the value
added of ﬁrms. 例如, 0%–50% means that the ﬁrm size would be less than half of the actual
ﬁrm size if all distortions were removed. 相似地, 200%+ means that the ﬁrm size would be more
than double without distortions. The entries are the shares of ﬁrms (averaged over the period).
来源: Authors’ calculations based on Government of Viet Nam, General Statistics Ofﬁce. Annual
Survey of Enterprises. https://www.gso.gov.vn/default_en.aspx?tabid=479&idmid=5

These results for Viet Nam are similar to those of the PRC, 印度, and the US in
Hsieh and Klenow (2009).24

24The Government of Viet Nam has launched various schemes to improve the performance of SMEs,
including credit funds and worker trainings (特兰, Le, and Nguyen 2008, PP. 347–59). 然而, unlike India, 在哪里
size-related policies are explicitly imposed by law, such policies in Viet Nam are only guidelines. We cannot identify
from the data which individual ﬁrms are eligible for support or have received any form of support. It is thus difﬁcult
for us to conduct an analysis similar to that of Hsieh and Klenow (2009).

112 ASIAN DEVELOPMENT REVIEW

数字 7. Distribution of Firm Size

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

笔记: The solid line indicates the actual size distribution, while the dashed line indicates the efﬁcient size distribution.
来源: Authors’ calculations based on Government of Viet Nam, General Statistics Ofﬁce. Annual Survey of
Enterprises. https://www.gso.gov.vn/default_en.aspx?tabid=479&idmid=5

乙.

Robustness check: different parameter values

One may be concerned that our analysis is sensitive to the choice of parameter
values and sample selection because our results are based on speciﬁc parameter
values such as σ = 3. To address this concern, we reconduct all the analyses using
different parameter values. Because it is tedious to examine all the results, 这
section examines (我) how sensitive the estimated TFPR and TFP gains (报道
in section IV and in Table 3) are to the choice of parameter values and sample
选择, 和 (二) the correlation between alternative and baseline TFPR. 在这个
robustness check, we report absolute TFP gains rather than relative TFP gains (到
the US) because we only change the parameter values in Viet Nam (not in the US).
We ﬁrst examine whether the results are sensitive to the value of the elasticity
of substitution, σ . In the baseline analysis, following Hsieh and Klenow (2009), 我们
set σ = 3. This implies that the markup is 1.5 (= 3/(3 - 1)). As a robustness check,
we set σ = 2 and σ = 6, and the corresponding markups are 2 (= 2/(2 - 1)) 和
1.2 (= 6/(6 - 1)), 分别. The second and third columns in Table 5 展示
结果. The TFP gains are somewhat sensitive to the value of the elasticity of
substitution. The TFP gains are 65.3% when σ = 2 和 161.9% when σ = 6, 尽管
the baseline TFP gains are 86.8%.25

25This result is consistent with equation (17), which implies that the TFP gains will be large if the elasticity

of substitution is large.

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 113

哦
t

e
v
我
t
A
我
e
右
y
t
我
v
我
t
C
你
d
哦
r
磷
r
哦
t
C
A
F

我
A
t
哦
时间
d
e
s
A
乙
–
e
你
n
e
v
e
右
G
n
我
z
我
我
A
你
q
乙
米
哦
r
F

s
n
我
A
G
y
t
我
v
我
t
C
你
d
哦
r
磷
r
哦
t
C
A
F

我
A
t
哦
时间

:
k
C
e
H
C
s
s
e
n
t
s
你
乙
哦
右

.
5

e
我
乙
A
时间

s
e
t
A
t
S
d
e
t
我
n
U
e
H
t
n
我

s
n
我
A
G
7
9
9
1

7
k
C
e
H
C

6
k
C
e
H
C

5
k
C
e
H
C

4
k
C
e
H
C

3
k
C
e
H
C

2
k
C
e
H
C

1
k
C
e
H
C

s
s
e
n
t
s
你
乙
哦
右

e
n
我
我
e
s
A
乙

)
8
(

)
7
(

)
6
(

)
5
(

)
4
(

)
3
(

)
2
(

)
1
(

s
e
t
A
t
S
d
e
t
我
n
U

C
ﬁ
我
C
e
p
s
米

r
我
F

米
A
氮

t
e
我
V

3
=
σ

%
2

6
8
1
0
1

8
6
0

5
4
6

8
4
9
0

%
2

3
6
2
7
9

1
7
0

7
5
7

5
9
9
0

%
1

7
4
9
,
0
0
1

4
9
7
.
0

1
6
.
0

0
.
0
4

%
1

2
3
8
,
0
0
1

9
8
8
.
0

4
6
.
0

0
.
8
6

3
=
σ

3
/
1

%
1

8
4
8
,
0
0
1

7
2
9
.
0

4
6
.
0

1
.
0
7

6
=
σ

2
=
σ

3
=
σ

s
e
t
A
t
S
d
e
t
我
n
U

%
1

2
1
6
,
0
0
1

9
7
.
0

9
.
1
6
1

4
9
9
.
0

%
1

1
0
6
,
0
0
1

7
9
9
.
0

8
7
.
0

3
.
5
6

%
1

1
0
6
,
0
0
1

0
0
0
.
1

9
7
.
0

8
.
6
8

右
磷
F
时间
e
n
我
我
e
s
A
乙
H
t
我

w
n
哦
我
t
A
我
e
r
r
哦
C

σ
:
y
t
我
C
我
t
s
A
我
乙

:
y
G
哦
我
哦
n
H
C
e
时间

)

右
磷
F
时间
(
D
S

米

我
r
时间

氮

)

(

s
n
我
A
G
磷
F
时间

d
我
乙
A
t
?
X
p
s
A
.
n
e
_
t
我
你
A
F
e
d
/
n
v
.
v
哦
G
.
哦
s
G
.
w
w
w

/
/
:
s
p
t
t
H

.
s
e
s
我
r
p
r
e
t
n
乙

F
哦

y
e
v
r
你
S

我
A
你
n
n
A

.
e
C
ﬁ
F
氧
s
C
我
t
s
我
t
A
t
S

我
A
r
e
n
e
G

米
A
氮

t
e
我
V

F
哦

t
n
e
米
n
r
e
v
哦
G
n
哦

d
e
s
A
乙

s
n
哦
我
t
A
我
你
C
我
A
C

’
s
r
哦
H
t
你
A

:
e
C
r
你
哦
S

5
=
d
我
米
d
我
&
9
7
4
=

.
y
t
我
v
我
t
C
你
d
哦
r
p

r
哦
t
C
A
F

我
A
t
哦
t
d
e
s
A
乙
–
e
你
n
e
v
e
r

=
右
磷
F
时间

,
y
t
我
v
我
t
C
你
d
哦
r
p

r
哦
t
C
A
F

我
A
t
哦
t

=
磷
F
时间

,
n
哦
我
t
A
我
v
e
d

d
r
A
d
n
A
t
s

=
D
S

.
3
e
我
乙
A
时间
米
哦
r
F
d
e
n
我
A
t
乙
哦
s
我

e
n
我
我
e
s
A
乙

e
H
时间

:
e
t
哦
氮

我
e
n
A
磷

d
e
C
n
A
我
A
乙

d
e
C
n
A
我
A
乙
n
U

e
r
你
t
C
你
r
t
s

我
e
n
A
磷

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

114 ASIAN DEVELOPMENT REVIEW

尽管如此, the estimated TFPR is qualitatively similar to the baseline
结果. 桌子 5 also reports the correlation with baseline TFPR, 这是 0.997
when σ = 2 和 0.994 when σ = 6. These high correlations suggest that the results
are quantitatively different from—but qualitatively similar to—the baseline results.26
The standard deviation of lnTFPR is 0.78 when σ = 2 和 0.79 when σ = 6, 两个都
of which are similar to that of the baseline model (0.79).

We also examine the sensitivity of the results to the value of the technology
parameter (capital share αs). We examine two different technologies. One is αs =
1/3, as in Ziebarth (2013), and the other is the Vietnamese cost share, 这是
deﬁned as the industry-year average capital share of the sample ﬁrms. The results are
presented in the fourth and ﬁfth columns in Table 5. The TFP gains are 70.1% 什么时候
αs = 1/3 和 68% when we assume Vietnamese technology. The correlation with
the baseline TFPR is 0.927 when αs = 1/3 和 0.889 when we assume Vietnamese
技术. The standard deviation of lnTFPR is 0.64 for both cases. 类似于
the value of the elasticity of substitution, the results are quantitatively different
from—but qualitatively similar to—the baseline results.

One may also be concerned that the technology parameter αs is heterogeneous
across ﬁrms even within industries. To address this concern, we use the ﬁrm-level
capital share so that the capital share can vary across ﬁrms.27 The results are
presented in the sixth column in Table 5 and are similar to the baseline results,
although the TFP gains are somewhat sensitive to the technology parameters. 这
TFP gains are 40%. The correlation with the baseline TFPR is 0.794. The standard
deviation of lnTFPR is 0.61. These results together suggest that our main messages
remain unchanged even when we use different values for the technology parameter.
Another concern may be that the data are not precise, and thus Vietnamese
ﬁrm-level data are subject to measurement error problems. Although we cannot
rule out arbitrary measurement error, we can try to gauge whether our results are
attributable to some speciﬁc forms of measurement error. We focus on two forms of
measurement error. 第一的, serious measurement error, possibly because of reporting
错误, tends to appear as outliers. We trimmed 2% from the tails (below the second
percentile and above the 98th percentile), 而不是 1% as in the baseline analysis,
and examined how sensitive the results are to the trim values. The seventh column
reports the results. The TFP gains are 75.7%. The correlation with the baseline
TFPR remains high at 0.995. The standard deviation of lnTFPR (0.71) is slightly
lower than that of the baseline model (0.79).

26It may also be important to allow the elasticities to vary across industries. Although Broda, Greenﬁeld,
and Weinstein (2006) estimated the elasticity of substitution for various economies, Viet Nam is not covered in their
分析. We leave this exercise for future research.

27Note that ξs can vary across ﬁrms if the capital share is different across ﬁrms (参见方程 [12]); 那是,
TFPR will not necessarily be proportional to the capital and output wedges. We thus present the results for reference
仅有的. From equation (11), if the technology parameter is heterogeneous across ﬁrms (作为 (= R Ksi /Psi Ysi ), distortions
appear only in τY si because τK si will be zero.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 115

We also estimate the TFP gains for ﬁrms that survived throughout the sample
时期 (balanced panel). This exercise enables us to control for the effects of ﬁrm
entry and exit. The eighth column presents the results. This exercise reduces the
sample size substantially (N= 10,186). 尽管如此, the estimated TFP gains are
large and the correlation with baseline TFP is high: 64.5% 和 0.948, 分别.
The standard deviation of lnTFPR is 0.68, which is comparable to that of the baseline
模型. The results suggest that about three-quarters of TFP gains come from the
incumbent ﬁrms, while the rest of the gains come from ﬁrms entering and exiting
the market. We can thus conclude that the results from the balanced panel are
qualitatively similar to the baseline results.

总共, the magnitude of the TFP gains are somewhat sensitive to the choice
of the values of parameters σ and α. 尽管如此, our main messages remain
unchanged even if we use different parameter values or employ different sample
selection criteria: the potential TFP gains from removing distortions in Vietnamese
manufacturing are large.

V. Concluding Remarks

This paper employed the Hsieh and Klenow (2009) framework to investigate
misallocation and productivity linkages in Vietnamese manufacturing during the
period 2000–2009 using ﬁrm-level data. Our study has four major ﬁndings. 第一的,
misallocation in Viet Nam is comparable to that in the PRC, 印度, and Thailand.
This result is consistent with the common knowledge that resources in developing
economies are not efﬁciently allocated.

第二, there would be substantial improvement in TFP if no distortions
存在. If Viet Nam hypothetically moved to “US efﬁciency,” its TFP would be
boosted by 30.7%. 第三, smaller ﬁrms tend to face advantageous distortions, 尽管
larger ﬁrms tend to face disadvantageous ones. 最后, the efﬁcient distribution of
ﬁrm size is more dispersed than the actual size distribution. This result implies that
Viet Nam’s policies may constrain its large and most efﬁcient producers and coddle
its smallest and least efﬁcient ones.

These ﬁndings have policy implications. The ﬁrst ﬁnding suggests that,
similar to other developing economies, resource misallocation caused by the
distortions seems to be an important issue in Viet Nam. The second ﬁnding states
that the potential productivity gains from removing distortions in Vietnamese
manufacturing are large. The result
implies that reallocation would lead to
productivity gains that can accelerate potential growth through improved interﬁrm
resource allocation. The last two ﬁndings together imply that Viet Nam’s policies,
as stated earlier, may constrain its largest and most efﬁcient producers and coddle
its smallest and least efﬁcient ones. This suggests that policy makers need to focus

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

116 ASIAN DEVELOPMENT REVIEW

more attention on the allocation of resources. An important question, 所以, 是
whether or not resources are being allocated to productive ﬁrms.

参考

Aw, B.-Y., X. 陈, 和M. J. 罗伯茨. 2001. Journal of Development Economics 66 (1): 51–

86.

巴赫, 时间. 2014. Misallocation and Growth in an Entry Liberalization Period. Paper presented at

the Royal Economic Society. 伦敦. 一月.

Banerjee, A. 五、, and E. Duﬂo. 2005. Growth Theory through the Lens of Development Economics.

在P. Aghion and S. 氮. Durlauf, 编辑. Handbook of Economic Growth. 爱思唯尔.

Bartelsman, E., J. Haltiwanger, 和S. Scarpetta. 2013. Cross-Economy Differences in
生产率: The Role of Allocation and Selection. American Economic Review 103 (1):
305–34.

Bellone, F。, 和 J. Mallen–Pisano. 2013. Is Misallocation Higher in France than in the United

状态? University of Nice-Sophia-Antipolis. Unpublished.

Broda, C。, J. Greenﬁeld, 和D. Weinstein. 2006. From Groundnuts to Globalization: A Structural
Estimate of Trade and Growth. NBER Working Paper 12512. 剑桥, 嘛: 国家的
经济研究局.

Busso, M。, L. Madrigal, 和C. Pages-Serra. 2012. Productivity and Resource Misallocation in

拉美. The B.E. Journal of Macroeconomics 13 (1): 903–32.

Calligaris, S. 2015. Misallocation and Total Factor Productivity in Italy: Evidence from Firm-Level

数据. Labour 29 (4): 367–93.

Camacho, A。, and E. Conover. 2010. Misallocation and Productivity in Colombia’s Manufacturing.

IDB Working Paper Series 6783. 华盛顿, 直流: Inter-American Development Bank.

de Vries, G. J. 2014. Productivity in a Distorted Market: The Case of Brazil’s Retail Sector. 审查

of Income and Wealth 60 (3): 499–524.

Dheera–Aumpon, S. 2014. Misallocation and Manufacturing TFP in Thailand. Asia-Paciﬁc

Economic Literature 28 (2): 63–76.

Eslava, M。, J. Haltiwanger, A. Kugler, 和M. Kugler. 2004. The Effects of Structural Reforms on
Productivity and Proﬁtability Enhancing Reallocation: Evidence from Colombia. 杂志
of Development Economics 75 (2): 333–71.

促进, L。, J. Haltiwanger, 和C. Syverson. 2008. Reallocation, Firm Turnover, 和
Efﬁciency: Selection on Productivity or Proﬁtability? American Economic Review 98 (1):
394–425.

Government of Viet Nam, General Statistics Ofﬁce. Annual Survey of Enterprises. https://万维网

.gso.gov.vn/default_en.aspx?tabid=479&idmid=5

——. Population and Employment. http://www.gso.gov.vn/default_en.aspx?tabid=467
Ha, D. 时间. T。, and K. Kiyota. 2014. Firm-Level Evidence on Productivity Differentials and Turnover

in Vietnamese Manufacturing. Japanese Economic Review 65 (2): 193–217.

—— 2015. Misallocation, Trade Liberalization, and the Global Financial Crisis: The Case of

Vietnamese Manufacturing. Keio University. 九月. Unpublished.

Hopenhayn, H. A. 2014. Firms, Misallocation, and Aggregate Productivity: A Review. Annual

Review of Economics 6 (1): 735–70.

Hosono, K., 和M. Takizawa. 2013. Misallocation and the Dynamics of Establishment. 金融的

审查 112 (1): 180–209 (in Japanese).

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

MISALLOCATION AND PRODUCTIVITY IN VIETNAMESE MANUFACTURING 117

Hsieh, C.-T., 和P. J. Klenow. 2009. Misallocation and Manufacturing TFP in China and India.

Quarterly Journal of Economics 124 (4): 1403–48.

—— 2013. Misallocation and Manufacturing TFP in China and India: Correction Appendix.

斯坦福大学. Unpublished.

Le´on–Ledesma, M。, 和D. Christopoulos. 2016. Misallocation, Access to Finance, and Public

Credit: Firm-level Evidence. Asian Development Review 33 (2): 119–43.

McMillan, M。, 和D. Rodrik. 2011. Globalization, Structural Change and Productivity Growth.
在米. Bacchetta and M. Jansen, 编辑. Making Globalization Socially Sustainable. 日内瓦:
World Trade Organization and International Labour Organization.

Melitz, 中号. J. 2003. The Impact of Intra-Industry Trade Reallocations and Aggregate Industry

生产率. Econometrica 71 (6): 1695–725.

Midrigan, 五、, 和D. 是. 徐. 2014. Finance and Misallocation: Evidence from Plant-Level Data.

American Economic Review 104 (2): 422–58.

国家经济研究局. NBER–CES Manufacturing Industry Database.

http://www.nber.org/nberces/

Nishimura, K., 时间. Nakajima, and K. Kiyota. 2005. Does the Natural Selection Mechanism Still
Work in Severe Recessions? Examination of the Japanese Economy in the 1990s. 杂志
of Economic Behavior and Organization 58 (1): 53–78.

Restuccia, D ., 和R. Rogerson. 2013. Misallocation and Productivity. Review of Economic

Dynamics 16 (1): 1–10.

Syverson, C. 2011. What Determines Productivity? Journal of Economic Literature 49 (2): 326–

65.

特兰, 时间. C。, X. S. Le, and K. A. 阮. 2008. Vietnam’s Small and Medium-Sized Enterprises
发展: Characteristics, Constraints, and Policy Recommendations. 在H. Lim ed.,
SMEs in Asia and Globalization (ERIA Research Project Report 2007–5). Jakarta: 经济的
Research Institute for ASEAN and East Asia (ERIA).

Waugh, 中号. 乙. 2010. International Trade and Income Differences. American Economic Review

100 (5): 2093–124.

World Bank. 2014. World Development Indicators. 华盛顿, 直流.
Ziebarth, 氮. L. 2013. Are China and India Backward? Evidence from the 19th Century US Census

of Manufactures. Review of Economic Dynamics 16 (1): 86–99.

附录. Derivation of Equation (24)

From equations (7), (8), 和 (9), actual ﬁrm size is written as

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Psi Ysi = P σ

s Ys P 1−σ
(西德:7)
(西德:6)
Psi
Ps

= PsYs

1−σ

= θsY

(西德:28)

(1 + τK si )作为
Asi (1 − τY si )

(西德:30)

(西德:29)

1−σ

(西德:3)

(西德:12) (西德:16)

(西德:17)作为

(西德:14)

1−σ

1 + τK s j

(西德:16)
1 − τY s j

As j

(西德:17)

(A-1)

118 ASIAN DEVELOPMENT REVIEW

Efﬁcient ﬁrm size is obtained when τK si and τY si are equalized within industry s
(例如, τK si = τK s and τY si = τY s). From equation (A-1), the efﬁcient ﬁrm size is
written as

si Y ∗
P ∗

= θsY ∗ Aσ −1
j Aσ −1

si(西德:13)

s j

From equations (A-1) 和 (A-2), 我们有
(西德:29)σ −1
(西德:7)σ −1

(西德:6)

(西德:28)

P ∗
si Y ∗
si
Psi Ysi

= Y ∗
是

Ys
Y ∗
s

(1 + τK si )作为
1 − τY si

(A-2)

(A-3)

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
d
e
v
/
A
r
t
我
C
e
–
p
d

我

F
/

3
3
2
9
4
1
6
4
3
6
6
2
A
d
e
v
_
A
_
0
0
0
7
4
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3
下载pdf