2. Theoretical background
We estimate a version of the Euler equation to test for the political pecking order in China. This simple model allows us to formalize the political pecking order hypothesis and consistently estimate financial constraints for different types of firms. The Euler equation characterizes a firm’s optimal investment path and relates it to marginal adjust-ment costs in adjacent periods. A credit constrained firm behaves as if it has a higher discount rate for a given level of today’s adjustment costs. Ceteris paribus, constrained firms will then substitute investment tomorrow for invest -ment today.
According to the political pecking order in China, firms are discriminated against in their access to credit bad on their ownership. To test for different ownership effects, we split our sample between private, state-owned and for -eign companies. In our estimation
we include ) in the cost
function to allow for correlation between the previous and actual investment decision. We expect the labor to capital ratio (L/K ) to proxy for the marginal product of profits with respect to capital as it
is conceivable that higher investment leads to a higher capital to labor ratio. Firm-
able of sign and significance is our measure of credit constraints. Given our assumptions, the empirical equation that we
ek to estimate has the following standard specification:
(1)
In the above equation, I denotes gross investment in fixed asts, L is the number of employees, K is the level of the real capital stock (proxied by total asts), ΔTU cor-responds to the change in turnover and CF stands for cash flow. The subscripts i , k and t denote the firm, industry and time p
eriod, respectively; and λk,t captures the ctor-time specific effects, εi,t is the error term. We measure
financing constraints by the nsitivity of investment to cash flow. We argue that the larger this nsitivity, the more constrained the firm is since it has to rely on its internal funds to finance its investment. In order to validate our hypothesis of the political pecking order , we would expect an insignificant coefficient on the cash flow indicator (β4) when the sample is restricted to state-owned or to foreign firms, but a positive and significant coefficient when pri -vate firms are considered since they rank lowest in the political pecking order. To study the contingency of credit constraints for Chine private firms we include variables measuring the prence of foreign investment and state-firms by province and by province/ctor and interact tho with our proxies for credit constraints.
We anticipate the nsitivity of investment to cash flow to be lower for private firms located in provinces/ctors where FDI is abundant. An opposite result would in con-trast point to a crowding-out effect. As far as the effect of state corporate prence is concerned, we expect that the higher the prence of the state-owned corporate ctor, the more crowding-out there will be and the stronger the credit constraints for private firms will be.
cooperate
3. Data
The data that we u are firm-level data originating from the Oriana data t. This databa contains detailed finan -cial information on contact information, activities, and ownership of more than 20,000 Chine firms in the time period of 1998 to 2005. Our data was collected by local governments bad on the Accounting Standards for Busi-ness Enterpris (ASBE) system promulgated by the Min-istry of Finance in 1992.
4. Investment Equation Estimates
Results for our baline specification in (1) are reported in Table 1. They are shown for different types of domestic private firms (col.1, 4 and 7), public SOE firms (col. 2, 5 and 8) and foreign firms (col. 3, 6 and 9). All regressions include time dummies defined at the two-digit ctor level to control for all shifts in investment demand or expectations due to changes in industry-level conditions (for example, industry-wide technology changes, industry demand shocks, or the entry of new firms).
We estimate successively our model with OLS, IV and
Sandra Poncet 1, Walter Steingress 2 and Hylke Vandenbussche 3
This paper us a unique micro-level data-t on Chine firms to test for the existence of a "political-pecking order" in the allocation of credit. Our findings are threefold. Firstly, private Chine firms are credit constrained while State-owned firms and foreign-owned firms in China are not; Sec-ondly, the geographical and ctoral prence of foreign capital alleviates credit constraints faced by private Chine firms. Thirdly, geographical and ctoral prence of state firms aggravates financial constraints for private Chine firms (“crowding out”). Therefore it ems that ongoing restructuring of the state-owned c-tor and further liberalization of foreign capital inflows in China can help to circumvent financial constraints and can boost the investment of private firms.
Keywords: Investment-cashflow nsi-tivity, China, firm level data, foreign direct investment JEL Codes: E22; G32
1 Introduction
Capital market imperfections are believed to be very prent in China. By law, the largest Chine banks, which were predominantly state banks, were until 1998 instructed not to lend to private firms. This was embedded in a deep political notion that private firms do not rank high in terms of political status. This “political pecking order” in the allocation of credit where private Chine firms are discri
minated against compared to state-owned firms should in principle have been alleviated since 1998. But casual evidence suggests that credit constraints for private firms still exist becau they are rooted in deep social and political factors (Huang, 2003).
Several macro studies have emphasized the detrimental effect of local government interference in capital alloca-tion in China 4, with only a few studies at the micro-level.5 The analysis in this paper is a micro-level study carried out on Chine firm-level data that extends the literature in veral dimensions. First, it offers an explanation for the conundrum of firm growth in China despite evidence of credit constraints. We look at the prence of foreign capital and how it can mitigate the financial discrimination experienced by Chine private firms. Our evidence sug -gests that foreign firms in China do not face credit con -straints indicating that they have superior legal status com-pared to private firms (Naughton, 2007). Alternatively they may be less dependent on the local financial system in China since they can rely on other sources to finance their growth. Our results suggest that the stronger the prence of foreign capital in a ctor or region, the lower the finan -cial constraints faced by Chine private firms operating in the same region and ctor. The prence of foreign capi-tal somehow allows Chine private firms to bypass both the financial and legal obstacles that they face at home. In addition, we also asss the extent to which the effec -tive
ness of financial constraints on private firms’ activity is contingent on state firms’ prence in the local econ -omy. Our results show that state prence aggravates credit constraints faced by domestic Chine firms which points at a "crowding-out" effect where stronger prence of state firms makes it more difficult for private firms to access capital.
Credit allocation in China: firm-level evidence
ment decided to gradually liberalize its regime for inward FDI. Foreign investment was encouraged through lower tax rates, fewer and simplified administrative and customs procedures and, most importantly, duty free import of components and suppliers (Naughton, 2007). Bad on this we would expect private firms to face significantly lower financing constraints in provinces that have a greater intensity of foreign direct investment. To e whether FDI alleviates financial constraints, we u the basic specifica-tion of equation (1) and include variables measuring the importance of foreign investment, both as a main effect and interacted with our proxy for credit constraints. Simi-larly, it can be argued that state-prence in a province may have the opposite effect and may aggravate credit con-straints for private Chine firms caud amongst others by banks preference to lend to state-owned enterpris. We will u three types of measures. A first t of indi-cators are province-level indicators of the abundance of foreign capital and of the relative size of the state corpo-rate c
tor: the ratio of FDI over GDP and the ratio of
finally firm-fixed effects to check the robustness of our results. The focus of our attention goes to the sign and magnitude of the coefficient on the lagged cashflow which is our measures of credit constraints. As conjectured, we find that private firms in China significantly rely on their cash flow to finance their investments which is evidence of credit constraints, while SOEs and foreign firms do not.
Since, the OLS estimates may be biad due to the endogeneity of the cash flow, our proxy for internal finance. In columns 4 to 6 we apply an IV technique to address this where we u the cash flow over asts in peri-ods t-2 and t-3 as instruments. The results go through be it with a weaker significance of the positive coefficient on the private firms, suggesting that the endogeneity of the cash flow is not too rious an issue. We systematically check the validity of our instruments with Sargan’s J-test of overidentifying restrictions and find that our choice of instruments is appropriate.
In columns 7 to 9, we include firm fixed effects to con-trol for all unobrved time-invariant variables. It also controls for the possibility of a correlation between a time-invariant component of the error and the regressors which would make the pooled OLS estimation inconsist-ent. The obtained firm-fix
ed effects results confirm the OLS estimates.The specification in column 7 for Chine private firms suggests that holding other factors constant, a 10% increa in the cashflow ratio CF/K of private firms rais investment by about 1.1%. Since the average investment rate over our sample is 15%, this would mean an addi-tional 1.1 percentage point increa which is economically significant.
Overall the results suggest that private Chine firms face vere financial constraints while we find no such constraints for state-owned and foreign enterpris. Our findings thus confirm the hypothesis of Huang (2003) that the Chine capital market is characterized by political pecking order bad on firms’ ownership type. This find-ing of discrimination against private firms by financial institutions is at odds with the obrvation that the firms are the engine role of growth in the Chine economy. 5. Contingency of the relationship between investment and cash flow
In this ction we shed further light on the circumstances under which financial distortions may not reprent an impediment to economic activity. We test two condition-ing factors of the effectiveness of the discrimination of private firms by financial institutions: (1) the role of FDI in funding the Chine corporate ctor and (2) the size of the state-owned corporate ctor.
At the beginning of the 1980s, the Chine govern-
Table 1. OLS, IV technique and Firm-fixed effects to test for credit constraints across ownership types
(1) (2) (3) (4) (5) (6) (7) (8) (9) Dependent variable: Investment over
lagged total asts Private SOE Foreign Private SOE Foreign Private SOE Foreign
OLS OLS OLS IV IV IV FE FE FE Lag dependent (investment divided 0.095*** -0.009*** 0.100*** 0.125*** -0.048 0.114*** -0.365*** -0.623*** -0.528*** by lagged total asts) i, t-1 (0.014) (0.002) (0.019) (0.025) (0.039) (0.020) (0.016) (0.024) (0.018) Change in turnover over asts i, t 0.021*** 0.108** 0.012*** 0.027*** 0.070*** 0.013*** 0.038*** 0.062*** 0.010***
(0.004) (0.047) (0.003) (0.007) (0.025) (0.003) (0.003) (0.005) (0.003) Employment over asts i, t -0.357 -1.489** -0.247*** -3.219*** -7.320** -0.598 -6.906*** -14.646*** -5.827***
(0.251) (0.617) (0.094) (1.002) (3.342) (0.543) (1.423) (2.793) (1.779) Cash flow over asts i, t-1 0.121***-0.088 0.028 0.109* 0.076 0.013 0.134*** 0.090 -0.019
(0.037) (0.159) (0.023) (0.058) (0.203) (0.019) (0.033) (0.087) (0.033) Cash flow squared over asts i, t-1 -0.149*** -0.062*** 0.016**
(0.025) (0.018) (0.007) Sector-year fixed effects yes yes yes yes yes yes yes yes yes Firm fixed effects no no no no no no yes yes yes Obrvations number 9229 5766 7316 4152 1607 1994 9229 5766 7316 R-squared 0.074 0.191 0.026 0.110 0.124 0.027 0.195 0.386 0.314 Cragg-Donald F statistic (weak
identification test): 727 210 1605
Sargan statistic (overidentification test of all instruments):
Chi-sq(1) P-val 0.084
(0.776)
5.956**
(0.0147)
3.077*
(0.0794)
Standard errors in parenthes.
* significant at 10%; ** significant at 5%; *** significant at 1%
In columns 4 to 6, Cash flow over asts i, t-1 is instrumented with Cash flow over asts i, t-2 and Cash flow over asts i, t-3.
Table 2. Estimation of Investment to Cash flow nsitivities depending on the
share of Foreign Direct Investment for private Chine firms
)7(
lula
)6(
)5(
)4(
)3(
)2(
)1(
**563.0-
***563.0-
1-t,i r a v t n e d n e p e d d e g g a L*-0.270***-0.367***-0.366***-0.365***-0.365***
)610.0(
)610.0(
)610.0(
)610.0(
)220.0(
)610.0(
)610.0(
)
s t e s s a l a t o t d e g g a l y b d e d i v i d t n e m t s e v n i(
***930.0***930.0***930.0***930.0***360.0***930.0
***830.0
t,i s t e s s a r e v o r e v o n r u t n i e g n a h C
)300.0(
)300.0(
)300.0(
)300.0(
)800.0(
)300.0(
)300.0(
***432.0***332.0***542.0***332.0***502.0***802.0
***431.0
1-t,i s t e s s a r e v o
w o l f h s a C
)840.0(
)840.0(
)740.0(
)540.0(
)770.0(
)450.0(
)330.0(
**779.6-
***609.6-
t,i s t e s s a r e v o t n e m y o l p m
E*-4.553*-7.609***-7.584***-7.301***-7.302***
)924.1(
)924.1(
)434.1(
)434.1(
)944.2(
)324.1(
)324.1(
**921.0-
***941.0-
t,i d e r a u q s s t e s s a r e v o
w o l f h s a C*0.178**-0.122***-0.112***-0.102***-0.102***
)030.0(
)030.0(
)720.0(
)620.0(
)980.0(
)720.0(
)520.0(
Province level share of fixed ast investment financed by foreign
investment0.252
)571.0(妇女节快乐英文
t,i Y S C e c r u o s
Interaction with cash flow i, t-1.492*
(0.876)
Province level FDI/gdp-0.001
)200.0(
t,i Y S C e c r u o s
310.0-
t,i w o l f h s a c h t i w
n o i t c a r e t n Ibest regards
(0.014)
Share of foreign enterpris in tangible asts-0.104**
source Oriana i, t(0.051)
Interaction with cash flow i, t-0.473***
(0.145)
Share of foreign enterpris in total asts-0.063
source Oriana i, t(0.058)
Interaction with cash flow i, t-0.521***
(0.155)
Share of foreign enterpris in turnover0.003
source Oriana i, t(0.048) Interaction with cash flow i, t-0.390***
(0.140)
Share of foreign enterpris in sales0.003 source Oriana i, t(0.048) Interaction with cash flow i, t-0.391***
(0.140)
s e y
s e y
s e y
s e y
s e y
s e y
s e y
s t c e f f e d e x i f r a e y-)s t i g i d2(r o t c e s
9229
9229
9229
9229
3027
9229
9229
s n o i t a v r e s b O
502.0
502.0
702.0
802.0
133.0
143.0
402.0
taughtd e r a u q s-R
Standard errors in parenthes
六年级毕业试卷* significant at 10%; ** significant at 5%; *** significant at 1%
indonesia
CSY: Chine Statistical Yearbook
Oriana: firm-level databa from which we aggregated variables like province/ctor tangible asts, total
asts, turnover and sales.
Almost all specifications in Table 2 suggest that FDI eas Chine private firms’ credit constraints, as com -pared to estimates from the specification including only CF/K , reproduced in column 1. The coefficients on the interaction terms, CF/K times our proxies for foreign cap-ital, which are almost all negative and significant for pri -vate firms, suggest that the prence of foreign firms reduces credit constraints. Hence, there is no evidence of crowding-out. Our results indicate that private firms located in a location/ctor where foreign capital is abun -dant and where the state ctor is low are more in a posi-tion to overcome the financial market inefficiencies caud by Chine economic institutions and policies. We thus identify that FDI is one mechanism that helps firms to overcome financial constraints. FDI brings in scarce capi -tal, eas financing constraints and spurs growth and investment of private firms.
In addition we look at the prence of the state-owned corporate firms which may also be a conditioning factor of the effectiveness of private firms’ credit constraints in China. In line with Huang (2003), we expect that it is more difficult for private firms, lowered-tiered on the political pecking order, to have access to credit, in provinces/c -tors where the relative size of the State-owned corporate ctor is high.
Table 3 reports the obtained results and confirms our predictions that private firms have a higher n
sitivity of investment to cash are more financially constrained, in locations/industries where the state prence is high. This micro-level evidence is coherent with macro-level findings of Guariglia and Poncet (2008) and Boyreau Debray and Wei (2005). Overall, our results support the conjecture of Boyreau-Debray and Wei (2005) that the state-owned banking ctor favors inefficient State Owned firms at the expen of private owned firms, which face financial constraints that hinder them to grow. The size of the state-owned corporate ctor appears to affect the extent to which private firms investment depends on inter -nal finance. Indeed, firms competing directly with numer -ous state-owned enterpris in the same province/industry suffer from higher constraints in their investment deci-sion.
Overall our findings allow us to predict the likely impact of the ongoing reforms inducing further liberaliza-tion and state firms restructuring on the economic dyna -mism of the Chine economy. We interpret our findings as evidence that credit constraints for private firms are likely to be mitigated by the growing importance of for-eign firms in the Chine economy as well as the ongoing decline of the state economic predominance.
Notes
[1] Centre d’Economie de la Sorbonne, Université Paris 1 and CEPII,
[2] Boston University, US.
[3] Corresponding author, Université catholique de Lou -vain, Department of Economics, Place Montesquieu 3, 1348 Louvain-la-neuve, Belgium ; T :+3210474137 ; F : +3210473945 ; hylke.vandenbussche@uclouvain.be
[4] Allen et al. (2005), Guariglia and Poncet (2008).
[5] Héricourt and Poncet (2008), Guariglia, Liu and Song (2008).
[6] Oriana datat is made available by Bureau van Dijk. It is constructed from Huaxia credit.
References
Allen, F ., J. Qian, and M. Qian (2005), “Law, Finance and Economic Growth in China,” Journal of Financial Eco -nomic, 77, pp. 57–116.
Boyreau-Debray G. and S.-J. Wei, (2005), “Pitfalls of a State-dominated Financial System: The Ca of China” NBER Working Paper 11214.
Guariglia A. and S. Poncet, (2008), Could financial distor -tions be no impediment to economic growth after all? Evidence from China", Journal of Comparative Eco -nomics, Volume 36, Issue 4, pp. 633-657.
Guariglia, A., X. Liu and L. Song (2008), “Is the growth of Chine firms constrained by internal finance”, mimeo.Héricourt, J., and S. Poncet (2008), “FDI and Credit Con -straints: firm level evidence from China”, forthcoming Economic Systems.
Huang, Y . (2003), Selling China. Cambridge University Press, pp. 207.
erar是什么意思Naughton, B. (2007), The Chine Economy: Transition and Growth. MIT Press, pp. 332.
employment in state-owned firms over total employment respectively. A cond t of measures, we will u the “share of fixed asts investment financed by foreign sources” as a proxy of foreign capital and the “share of fixed asts investment financed by the state budget” as a proxy for state prence. A third t of indicators meas-ures the importance of the foreign and state ctor. Our purpo is to analyze their interactions with the cashflow variable in our baline specification in (1). Table 2 reports results on all the parate indicators that proxy for the abundance of foreign capital at the provincial and province/ctor level. Our wide lection of indicators allows us to account for
different aspects of the foreign prence and to test the robustness of our results.
Table 3. Estimation of Investment to Cash flow nsitivities depending on the share of State
Owned firms per industry for private Chine firms .
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Lag dependent i, t-1
-
0.365***-0.366***-0.365***-0.364***-0.364***-0.364***-0.364***(investment divided by lagged total asts)(0.016)(0.016)(0.016)(0.016)(0.016)(0.016)(0.016)Change in turnover over asts i, t
0.038***0.038***0.038***0.039***0.039***0.039***0.039***(0.003)(0.003)
(0.003)(0.003)(0.003)(0.003)(0.003)Cash flow over asts i, t-10.134***0.097**
-0.0950.096**0.080*0.086**0.085**(0.033)(0.044)
(0.162)(0.041)(0.041)(0.040)(0.040)Employment over asts i, t -6.906***-6.932***北京日语
-6.988***-6.994***-7.033***-7.089***-7.091***(1.423)(1.423)
(1.425)(1.424)(1.424)(1.425)(1.425)Cash flow over asts squared i, t -0.149***-0.138***
-0.129***-0.149***-0.149***-0.148***-0.148***(0.025)
(0.026)
(0.028)
(0.025)
(0.025)
(0.025)
(0.025)
Province level state share of employment -0.150)001.0(t ,i Y S C e c r u o s 3
83.0t ,i w o l f h s a c h t i w n o i t c a r e t n I (0.266)
Province level share of investment financed by state budget 0.071)680.0(t
,i Y S C e c r u o s 786.0t
报考mba的条件,i w o l f h s a c h t i w n o i t c a r e t n I (0.574)
Share of foreign enterpris in tangible asts -0.041)630.0(t
,i a n a i r O e c r u o s 712.0t
,i w o l f h s a c h t i w n o i t c a r e t n I (0.140)
Share of foreign enterpris in total asts -0.036)340.0(t
,i a n a i r O e c r u o s Interaction with cash flow i, t
0.336**(0.155)
Share of foreign enterpris in turnover 0.029source Oriana i, t
(0.044)Interaction with cash flow i, t 0.346**(0.173)
Share of foreign enterpris in sales 0.028source Oriana i, t
(0.044)Interaction with cash flow i, t 0.350**(0.173)ctor (2 digits)-year fixed effects yes yes yes yes yes yes yes 9229922992299229922992299229s n o i t a v r e s b O 5
02.05
02.05
02.04
02.05
02.01
43.04
02.0d
e r a u q s -R Standard errors in parenthes
* significant at 10%; ** significant at 5%; *** significant at 1%CSY: Chine Statistical Yearbook
Oriana: firm-level databa from which we aggregated variables like province/ctor tangible asts, total asts, turnover and sales.
