Major Apparel Exporting Countries in Asia: A Study of Technical Efficiency and Its Determinants

Abstract

Asia has many significant players in the global apparel market, contributing more than half of the apparel output. There is stiff competition between eight countries within the region to capture the market share—China, Bangladesh, India, Pakistan, Sri Lanka, Vietnam, Cambodia and Indonesia. Using raw data from the World Bank Enterprise Surveys, a two-stage analysis was conducted for each country—in the first stage efficiency scores were computed using data envelopment analysis, and in the second stage factors influencing efficiency were identified using tobit regression and disaggregate comparison. The results revealed a poor efficiency performance across countries, but the drivers varied between them. It was observed that smaller firms performed better and there was a need to revamp quality certification and training modules to promote competitiveness.

Keywords

Apparel manufacturing Asia data envelopment analysis technical efficiency tobit regression

Introduction

Apparel manufacturing has historically been a major sector in the global market and sees extensive participation from developing economies. Asia has emerged as a prominent region for the global apparel market because of certain favorable factors (Bhavani & Tendulkar, 2001; Ramaswamy & Gereffi, 1998). The World Bank put the spotlight on apparel manufacturing and exports, particularly the South Asian participants in the global market through a report comparing four leading South Asian apparel-producing countries (Bangladesh, India, Pakistan, and Sri Lanka) with their three closest competitors in South East Asia (Vietnam, Cambodia, and Indonesia) (World Bank, 2016).

After the removal of quotas and opening of apparel trade in 2005 (Ramaswamy & Gereffi, 1998), in just a decade more than half of the world apparel exports were attributed to these eight Asian countries—China, Bangladesh, India, Pakistan, Sri Lanka, Vietnam, Cambodia, and Indonesia (Table 1).

Table 1.

Leading Apparel Exporting Countries of Asia.

Country	World Clothing Exports (% Share)		Country Classification
Country	2015	2005	Country Classification
China	36.93	25.32	Upper middle income
Bangladesh	5.64	2.35	Lower middle income
Viet Nam	4.98	1.60	Lower middle income
India	3.87	2.98	Lower middle income
Indonesia	1.61	1.69	Lower middle income
Cambodia	1.26	0.75	Lower middle income
Pakistan	1.07	1.23	Lower middle income
Sri Lanka	1.01	0.98	Lower middle income
Total	56.37	36.91

Source: Authors’ calculations and compilation based on data from WTO & World Bank.

The South Asian Countries together account for more than one-tenth of the global apparel exports and are all developing countries or in the lower middle-income group. The apparel sector has relatively lower skill requirement and is a suitable employment avenue for the growing workforce. Overall, it has the highest female labor participation rate in industrial sectors. Increased employment is expected to improve the female labor participation rate for South Asia, which is currently stagnant with a share of around one-third. This is also expected to have a positive impact on the efficiency and competitiveness of the industry (Hill & Kalirajan, 1993).

China has been the largest apparel producer and exporter for a decade but is vulnerable with rising costs. This presents a great opportunity to the South Asian countries. The potential gains for India are immense—the estimate is that a 10% increase in China’s apparel prices has the potential to lead to creation of 1.2 million fresh jobs in India’s apparel industry. Moreover, the focus on safety and, social and ethical compliance has increased ever since incidents of massive fire in Bangladeshi garment factories. However, the implementation of ethical and social codes and standards entails significant financial investment as well as changes in management style that are taxing and difficult for small firms. The core of compliance is based on the supplier’s fear of losing business and power of the buyer. India has relatively better working conditions as compared to Bangladesh in the formal apparel manufacturing sector (Lopez-Acevedo & Robertson, 2016).

India needs to take swift action to actualize the benefits in the global apparel sector and ensure expansion of both exports and domestic sales—by increasing product diversity, improving productivity, improving market diversity, and shortening lead times (Lopez-Acevedo & Robertson, 2016). Most Indian firms are operating with excess production capacity and need to improve resource and capacity utilization to boost competitiveness (Hashim, 2005; Joshi & Singh, 2012). There is a lot of variation in interfirm performance and gains can be materialized by leveraging the correct factors (Bheda et al., 2001; Hill & Kalirajan, 1993, Kapelko & Lansink, 2014).

However, South East Asian countries are strong competition for the South Asian countries. They have shown better performance “in terms of overall apparel export performance, product diversity, and noncost related factors important to global buyers” (World Bank, 2016). But presence of firm-level inefficiencies may pull down competitiveness (Vixathep & Matsunaga, 2012b).

Although the specific policy needs are based on the nature of industry and challenges in the domestic economic environment, there is an urgent need for each country to focus on adopting policies that ease market access, reduce barriers to import of inputs, enhance logistics for export, and promote foreign investment (Lopez-Acevedo & Robertson, 2016). Although export participation for small firms in developing countries may not be adequate to improve productivity, it is seen that productive firms may take to exporting and have higher market participation (Deshmukh & Pyne, 2013). Further, it is observed that firms with low productivity do not sustain and exit the export market (Mallick & Yang, 2013). Since changes in technology take time to translate into productive gains, improvements in efficiency are imperative to enhance productivity. The firm-specific characteristics must be considered in the formulation & implementation of policies for performance improvement (Ngo et al., 2020). The policies must promote specific firm-level factors to improve “value-added productivity” and focus on enhancing competitiveness (Satpathy & Mishra, 2019). Efficiency is the key to sustaining and expanding export performance. “South Asia has taken many steps in recent years to support the textile and apparel sector, but it now needs to step up its game by tackling inefficiencies that are undercutting its competitiveness” (World Bank, 2016).

This article has two objectives: first, to examine the technical and scale efficiency of leading apparel exporting countries in Asia, and second, to study the factors influencing pure technical efficiency in these countries. The findings will aid policy design to increase efficiency, enhance competition, and improve performance. The article is organized into sections as follows: The next section describes the empirical framework, followed by the section on the analysis and discussion. The last section concludes this article.

Empirical Framework

Estimating Technical Efficiency—Data Envelopment Analysis (DEA)

Production efficiency refers to the ability to manufacture a given output with minimum inputs. Charnes et al. (1978) expanded on Farrell’s concept (1957) to introduce the methodology of DEA. DEA is a nonparametric technique for estimation, which makes use of linear programming to identify a production frontier such that all decision-making units or firms under consideration lie on or below this frontier. Firms on the frontier are taken as efficient and those lying below as inefficient, facilitating benchmarking. DEA is most suited to measure a firm’s efficiency where there are multiple inputs involved. The DEA approach has some distinct advantages over the other popular approaches of growth accounting and stochastic frontier approach (Grifell-Tatje & Lovell, 1995; Joshi & Singh, 2010; Kumar, 2006) and is a great tool where firms have varying technology and are in different stages of their life cycle (Van Biesebroeck, 2007). However, conventional hypothesis testing cannot be conducted with DEA and there is no accounting for noise (Coelli et al., 2005).

The model proposed by Charnes, Cooper, and Rhodes assumes constant returns to scale (CRS). Banker et al. (1984) proposed that the CRS model can be extended to variable returns to scale (VRS) model by adding the convexity constraint to it. This allows for decomposition of technical efficiency in the form of pure efficiency and scale efficiency. Input orientation is considered suitable as there are constraints on demand that are beyond the control of a firm or industry (Sharma et al., 2010). Let us take N firms whose efficiency is to be compared. The following input-oriented Banker, Charnes, and Cooper model provides efficiency measure of the ith reference firm:

Subject to \begin{matrix} M i n_{θ, λ} θ_{i} \\ Y λ \geq Y \\ X λ \leq θ_{i} X_{i} \\ N' λ = 1 \\ λ \geq 0 \end{matrix}

where i_i provides the efficiency score for ith firm; Yi and Xi represent the output and input vectors of ith firm, respectively; Y and X are the output and input matrices for all the N firms; N' represents the unit vector and λ is an N × 1 vector of constants.

The firm is considered technically efficient and lies on the efficiency frontier if and only if the optimal value of θ_i is equal to one. A value <1 indicates relatively inefficient firm lying below the frontier. For the minimum number of decision-making units required for discrimination between efficient and inefficient performers, there is no fixed rule of thumb, three times the number of inputs is acceptable (Sarkis, 2007). DEA has been used extensively to study efficiency in apparel sector (Gambhir & Sharma, 2017; Joshi & Singh, 2012; Mok et al, 2010; Vixathep & Matsunaga, 2012a, 2012b). Given that technical efficiency is to be measured using traditional input–output variables, the widely used and time-tested Banker, Charnes, and Cooper model is applied under input orientation.

Identifying Influencing Factors—Tobit Regression and Disaggregate Analysis

To identify firm characteristics or attributes that influence technical efficiency for apparel firms Tobit regression and disaggregate analysis are used. Regression uses observed values of independent variable(s) to estimate or predict corresponding values of the dependent variable, both metric in nature (Cooper et al., 2012). The basic bivariate model is given as

Y = β_{0} + β_{1} X i

where Y is the dependent variable, Xi the independent variables, β₀ the constant, and β₁ the regression coefficient.

The efficiency scores computed using DEA in the first stage are regressed on selected independent variables to identify the drivers of technical efficiency. Tobit regression is a censored regression model that is suitable where the dependent variable has its value constrained, such as the efficiency scores that are constrained between 0 and 1. Tobit model has been used widely for identifying determinants of efficiency and is the selected model for analysis (Joshi & Singh, 2012). Disaggregate analysis compares performance of subgroups based on certain categorical firm-level attributes (Bheda et al., 2001; Chapelle & Plane, 2005; Kundi & Sharma, 2015).

Data and Variables

The efficiency analysis is conducted for the eight major Asian apparel-producing countries—China, Bangladesh, India, Pakistan, Sri Lanka, Cambodia, Vietnam, and Indonesia. Based on geographical coverage and income, all the South Asian and South East Asian countries selected for the study were classified as “Lower Middle Income,” whereas China was designated as “Upper Middle Income” (World Bank, 2017).

The raw data for relevant variables (Table 2) were taken from the recent Enterprise Surveys conducted by the World Bank in the respective countries (Enterprise Surveys, 2016); considered a reliable data source (Chapelle & Plane, 2005; Waldkirch, 2014).

Table 2.

Data Source and Scope for Country-Wise Efficiency Analysis.

Region/ Satus	Country	Primary Fiscal Year	Year of Survey Report	Number of Garment Firms Surveyed	Number of Firms Eligible for Efficiency Analysis
World Leader	China	Jan to Dec 2011	2012	116	84
South Asia	Bangladesh	2011–12	2013	196	170
South Asia	India	2012–13	2014	198	138
South Asia	Pakistan	2011–12	2013	81	15
South Asia	Sri Lanka	2010–11	2011	107	59
South East Asia	Cambodia	Jan to Dec 2015	2016	46	44
South East Asia	Indonesia	Jan to Dec 2014	2015	118	100
South East Asia	Vietnam	Jan to Dec 2014	2015	145	98

Source: Enterprise Surveys & Authors’ Analysis.

Output is taken as annual sales for the fiscal year, capital is the net book value of machinery, vehicles, and equipment for the fiscal year, and labor is the annual cost of labor, Energy is the annual cost of fuel and electricity, and materials is the annual cost of raw material, and so on. It is not necessary to deflate the financial figures in a cross-sectional efficiency analysis. Firm-level characteristics said to have influence on efficiency performance are firm age, export orientation, capital intensity, and scale of operation (Chapelle & Plane, 2005; Hashim, 2005; Hill & Kalirajan, 1993; Joshi & Singh, 2012; Mok et al, 2010; Vixathep & Matsunaga, 2012a; Vixathep & Matsunaga, 2012b; Wadud, 2004). Based on the data set, relevant variables for age, export intensity, capital intensity, and labor-staff ratio were considered under tobit analysis (Table 3).

Table 3.

Variables Used for Tobit Regression.

Nature	Variable	Description	Scale	Expected Relation
Dependent Variable	VRSTE	y*: Pure Technical Efficiency	Ratio
Independent	Age	A: Age of the firm from incorporation to study year	Ratio	+
Independent	Export Intensity	EI: Exports as % of Total Sales	Ratio	+
Independent	Capital Intensity	CI: Capital/Sales	Ratio	+/–
Independent	Labour-Staff Ratio	LSR: Ratio of Number of workers to Number of staff	Ratio	+

Source: Enterprise Surveys & Authors’ Selection.

The equation for the proposed Tobit model is

y *_{m} = α + β_{1} A_{m} + β_{2} E I_{m} + β_{3} C I_{m} + β_{4} L S R_{m} + ε_{t m}

where m represents the firm; β₁, β₂, β₃, β₄ represent parameters/coefficients to be estimated; and ε_t is the error term.

For the disaggregate analysis, three categorical attributes were considered as defined by the World Bank.

Firm Size: Large firms have at least 100 employees and MSME firms have <100 employees.

Quality Certification: If the firm has any international quality certification, such as ISO 9000 for quality management and assurance) or ISO 14000 for environmental management standards, or Hazard Analysis and Critical Control Points for food safety clothing.

Training Conducted: If the firm conducted or participated in any training for workers and employees in fiscal year.

Analysis and Discussion

Efficiency Analysis: Country-Wise Empirical Results

There is a wide variation in values of descriptive statistics between the countries as well as for the variables within a country (see Appendix). The efficiency analysis revealed that all countries have a poor efficiency performance in the apparel sector in terms of the average scores for technical efficiency (CRSTE), pure technical efficiency (VRSTE), and scale efficiency (SE) (Table 4).

Table 4.

Country-Wise Average Efficiency Scores.

Region/Status	Country	CRSTE	VRSTE	SE
World leader	China	0.142	0.338	0.553
South Asia	Bangladesh	0.45	0.594	0.761
South Asia	India	0.252	0.427	0.648
South Asia	pakistan	0.583	0.769	0.747
South Asia	Sri lanka	0.256	0.496	0.512
South East Asia	Cambodia	0.401	0.571	0.767
South East Asia	Indonesia	0.013	0.236	0.064
South East Asia	Vietnam	0.347	0.543	0.667

It is evident that majority of the firms for each country are inefficient in terms of CRSTE, VRSTE, and SE (Table 5). However, further segregation of inefficient firms shows that there significant proportion of firms are closer to efficiency with a good scope for improvement.

Table 5.

Country-Wise Distribution of Firms Based on Efficiency Performance.

Region/Status	Country	Distribution of Firms (in Percentage)
		CRSTE			VRSTE			SE
		EFF	CTE	PP	EFF	CTE	PP	EFF	CTE	PP
World leader	China	3.57	2.38	94.05	10.71	11.9	77.38	3.57	41.67	54.76
South Asia	Bangladesh	9.41	11.76	78.82	17.65	27.06	55.29	9.41	62.94	27.65
South Asia	India	2.9	3.62	93.48	13.04	9.42	77.54	3.62	55.8	40.58
Ssouth Asia	Pakistan	40	6.67	53.33	60	13.33	26.67	40	26.67	33.33
South Asia	Sri lanka	10.17	3.39	86.44	22.03	6.78	71.19	10.17	27.12	62.71
South East Asia	Cambodia	11.36	11.36	77.27	18.18	25	56.82	11.36	61.36	27.27
South East Asia	Indonesia	1	0	99	9	8	83	1	4	95
South East Asia	Vietnam	14.29	6.12	79.59	26.53	12.24	61.22	15.31	42.86	41.84

Note: EFF—Efficient (Value 1); CTE—Closer to Efficiency (0.6 <– value < 1); PP – Poor Performers (value < 0.6).

Almost all countries have majority firms experiencing increasing returns to scale (IRS) except Bangladesh (Figure 1). The operational scale must be overhauled to benefit from scale economies and improve efficiency performance.

Figure 1.

Country-Wise Firm Distribution based on Returns to Scale.

China

Despite being number one in the apparel market, China has a dismal efficiency situation. More than three-fourths of the firms are PP with VRSTE scores <0.6. Firms that are CTE in terms of SE constitute 41.67%. It is seen that 75% of firms show IRS; firms can enhance efficiency by scaling to the optimal plant size. Most firms have a material slack and the mean slack value is highest for materials implying gross material mismanagement or wastage.

South Asia

South Asia as a region also reveals dismal efficiency performance with huge variation in average efficiency scores for CRSTE, VRSTE, and SE. Only India and Sri Lanka have shown a similar performance in their respective study periods. India is a laggard in terms of pure efficiency with more than three-fourth of the sample firms revealed as PP on VRSTE. There is immense scope and necessity for efficiency improvement in South Asia, particularly for India.

For Bangladesh, less than 10% of the firms are efficient overall. It is performing best in terms of SE, with around two-thirds of the firms CTE. Around 70% of the firms are showing decreasing returns to scale. This implies that making corrections and moving to optimum scale holds more efficiency benefits for Bangladeshi apparel sector. The most mismanaged input is energy with 21.76% firms having slack on it, whereas materials input is most efficiently managed. The focus should be on improving the availability and utilization of energy in its apparel sector.

For India, the situation is alarming and needs immediate focus and redressal. There is widespread inefficiency—both in terms of pure technical component and the scale component, with just 2.90% efficient firms overall. Around 77.54% firms are PP on VRSTE while 40.58% firms are PP on SE. More than half of the firms are CTE in terms of SE, and 73.91% firms show IRS. Any change or improvement in scale is likely to improve efficiency faster for these firms. Capital and energy are most mismanaged in Indian apparel sector. There is an urgent need to focus and improve VRSTE and labor needs to be trained and motivated to improve utilization of other inputs.

For Pakistan, the small sample size is a constraint, which is making a general inference. In terms of VRSTE, only 26.67% PP. More than half the firms show IRS, capital is the most mismanaged input, followed by labor and materials. No firm is showing slack on energy input.

For Sri Lanka, it is seen that 71.19% firms are PP in terms of VRSTE, whereas 62.71% are performing poorly on SE. More than three-fourths of the firms show IRS. Almost all the inputs are largely mismanaged in Sri Lankan apparel sector. Sri Lankan enterprises must work on improving resource utilization and enhancing the scale of production.

South East Asia

South East Asian region also shows poor efficiency performance despite growth in terms of output, employment, and exports. Cambodia and Vietnam show similar average efficiency scores for CRSTE, VRSTE, and SE. Indonesia has extremely low values of efficiency indices on average. South East Asian region needs to urgently focus on efficiency component to sustain the stiff competition it gives to South Asia in the long run.

Although Cambodia has performed best within South East Asia, less than one-fifth of the firms are efficient in terms of VRSTE, with 56.82% revealed to be PP. About two-third of the firms are CTE for SE and 86.36% of firms show IRS.

Indonesia has an extremely dismal performance. Overall 99% firms are PP on efficiency and about 95% of firms show IRS. Labor is the most mismanaged input in Indonesian apparel sector with 15% firms showing slack on it. Indonesia needs to focus on overhauling production performance and efficiency urgently and manage workers in a better manner.

Vietnam has only around one-fourth of the firms efficient in terms of VRSTE, and 61.22% firms as PP. In terms of SE, 42.86% are CTE and 72.45% firms have IRS. The analysis of input slacks shows that a negligible number of firms show slack on capital and labor. Energy and material seem to be efficiently managed.

Suggestions for Efficiency Improvement

An interesting snapshot emerges when we study the different efficiency results of all eight countries (Table 6). South Asian countries have shown better efficiency performance. The poorest efficiency performance is seen in Indonesia, considered to be a strong competitor from South East Asia. Such dismal efficiency parameters raise a doubt on its continuance as a strong competitor.

Table 6.

Country-Wise Summary of Efficiency Results.

Region/ Status	Country	Input Slacks		Major Returns to Scale	Inefficiency (Dominant)		Action for Efficiency Improvement
Region/ Status	Country	Max. Firms Slack on	Min. Firms Slack on	Major Returns to Scale	CTE	PP	Action for Efficiency Improvement
World leader	China	Materials	Energy	IRS	SE	VRSTE	Improve scale of production; Efficient materials utilization
South Asia	Bangladesh	Energy	Materials	DRS	SE	VRSTE	Scale down to productive plant size, Focus on Energy access and utilization
South Asia	India	Capital	Labour	IRS	SE	VRSTE	Improve scale of production; Focus on Capital utilization and upgradation
South Asia	Pakistan	Capital	Energy*	IRS	SE	SE	Improve scale of production; Focus on management of Capital input
South Asia	Sri lanka	Capital	Materials	IRS	SE	VRSTE	Improve scale of production; Focus on management of Capital input
South East Asia	Cambodia	Labour	Capital	IRS	SE	VRSTE	Improve scale of production; Focus on performance of labour
South East Asia	Indonesia	Labour	Energy	IRS	VRSTE	SE	Improve scale of production; Focus on performance of labour and technical efficiency
South East Asia	Vietnam	Capital	Energy; Materials	IRS	SE	VRSTE	Improve scale of production; Focus on management of Capital input

Although we cannot directly compare the efficiency scores across countries as they relate to different years, we can identify the focal improvement areas for each country. It is observed that apparel manufacturing sector in all countries except for Bangladesh show IRS, consistent with the small scale of operations in these regions (D’Souza, 2016). South Asian countries need to focus most on managing the capital, primarily plant, and machinery. The focus should be on upgradation as well as optimum use of existing facilities. Innovation is the key to sustainable growth (Joshi & Singh, 2010). Bangladesh needs to focus additionally on energy input which includes timely access to and efficient utilization of fuel and energy.

For South East Asian countries, training the workers, enhancing skills, and promoting labor productivity needs most attention. Across the countries, apart from enhancing SE there is an urgent need to focus on the VRSTE. The suboptimal resource utilization will reduce competitiveness and stall export gains unless addressed.

Factors Influencing Pure Technical Efficiency: Country-Wise Results

The results of the Tobit regression show that the coefficients for age were significant for Bangladesh, India, and Indonesia (Table 7). However, only India showed an expected positive relation with older firms showing better efficiency. For Bangladesh and Indonesia, the coefficients were negative implying that younger firms have better efficiency. Contrary to expectation, export intensity showed negative and significant relation for Sri Lanka and Vietnam. This variable was not significant for all the remaining countries. This implies that while exporting process may influence the efficiency the volume of exports may not be a significant driver of efficiency. This result seems to support the self-selection hypothesis that is the directionality may flow in reverse with efficient firms being the ones present in export market (Haidar, 2012).

Table 7.

Results of Tobit Regression.

Country Under Study	Dependent Variable−VRSTE (Pure Technical Efficiency)										Number of observations
	Independent Variables
	Age		Export intensity		Capital intensity		Labor-staff ratio		Constant
	Coefficient	Standard Error	Coefficient	Standard Error	Coefficient	Standard Error	Coefficient	Standard Error	Coefficient	Standard Error
China	−0.0001	0.0002	0.0012	0.0009	−0.1219	0.0858	−0.0025	0.0025	0.4561^***	0.1012	83
BAN	−0.0070^**	0.0030	0.0001	0.0006	−0.0148	0.0145	−0.0009	0.0009	0.7864^***	0.0820	169
India	0.0061^**	0.0026	−0.0006	0.0006	−0.0053	0.0035	−0.0023	0.0018	0.4597^***	0.0818	138
PAK	−0.0001	0.0192	0.0035	0.0038	−0.0005	0.0006	0.2281	0.1422	0.1516	0.7036	15
SL	0.0020	0.0032	−0.0030^**	0.0012	−0.0769^*	0.0384	−0.0108	0.0121	0.7106^***	0.1302	58
CAM	−0.0180	0.0109	−0.0001	0.0011	0.0054^**	0.0021	0.0039^**	0.0018	0.6387^***	0.1452	44
IDSA	−0.0084^**	0.0038	0.0001	0.0011	0.0247	0.0205	0.0130^***	0.0042	0.2658^***	0.0894	99
VTM	0.0074	0.0048	−0.0023^**	0.0010	0.0008	0.0206	0.0037	0.0041	0.6104^***	0.0953	97

Note: Significant at *10%, **5%, ***1%. BAN—Bangladesh; PAK—Pakistan; SL—Sri Lanka; CAM—Cambodia, IDSA— Indonesia and VTM—Vietnam.

The coefficient for capital intensity was significant for only Sri Lanka and Cambodia. For Sri Lanka, it showed a negative relation with lower capital intensity leading to more efficiency while in Cambodia it is positive. The coefficient for labor-staff ratio was expectedly positive and significant for only Cambodia and Indonesia implying that higher labor per staff ratio is positively linked with pure technical efficiency. However, this variable was not significant for the remaining countries.

It has been observed that there is regional variation in productivity and efficiency performance (Kumar, 2006). The results indicate that even the drivers and the relations may differ between countries within a geographical region. Each country needs to develop a unique and tailor-made support programme for improving efficiency of apparel manufacturing firms according to the local business environment. It is dangerous to imitate or adopt a policy without paying attention to the individual requirements of the sector in a country.

The disaggregate analysis showed that the VRSTE score on average is higher for MSME firms supporting the proposition that small scale leads to better resource management. The small scale is responsible for lower SE scores on average as well.

For China, India, Pakistan, Sri Lanka, and Vietnam, most of the sample firms did not have any form of quality certification and firms without international quality certification had better technical efficiency. This is contrary to expectation as quality certification and the technical efficiency are generally considered to be linked (Gambhir & Sharma, 2017). There seems to be a possibility that the requirements for the quality certification may not promote efficiency in production. This may also be due to the specific type of certification obtained by the sample firms and should be investigated in future studies. None of the sample firms for Cambodia had any quality certification. Only in the case of Bangladesh and Indonesia it was seen that firms with quality certification were performing better in terms of technical efficiency. The countries would do better by promoting quality certification that considers the actual inputs and production techniques to improve the quality of the final products leading to enhanced competitiveness. There is a need to cultivate an environment where all processes and business decisions are driven by quality and value addition.

Bangladesh, India, Pakistan, Cambodia, and Indonesia reported higher VRSTE scores on average for firms that conducted training programs in the financial year. Few firms had engaged in training in Pakistan, Sri Lanka, Cambodia and Indonesia. Only China, Sri Lanka, and Vietnam had firms with no training conducted showing higher efficiency. Amongst this group, only China has a great majority of firms with training programs. This result is contrary to expectation and may have resulted from mismatch of the training module and the requirements of the firms.

Conclusion

Asian countries have emerged as leading suppliers in the global apparel industry. The efficiency performance of apparel manufacturing sector in China, South Asia, and South East Asia has been studied using DEA, under input orientation and VRS. Designing and adopting suitable promotional policies for the sector entails studying the impact of firm-level variables. The key findings are as follows: First, there is a dismal efficiency situation in all the countries evaluated and an urgent need for focusing on improving pure technical efficiency. Second, there is a need to promote efficient of resource utilization, and improve skills and management, while moving to the optimum plant scale and capacity. Third, the factors that influence the pure technical efficiency of firms vary between countries and include age, export intensity, capital intensity, and labor-staff ratio. The nature of the relation also varies between countries. Fourth, the disaggregate analysis showed MSME firms as better performers and highlighted that quality certifications and training modules must be designed in a manner to promote efficiency. By tapping the specific firm factors, competitiveness can be enhanced leading to higher efficiency and sustainable growth.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Appendix

References

Banker

R. D.

, Charnes

, & Cooper

W. W.

(1984). Some models for estimating technical and scale efficiencies in data envelopment analysis. Management Science, 30(9), 1078–1092. http://astro.temple.edu/~banker/DEA/13%20Some%20Models%20for%20Estimating%20Technical%20and%20Scale%20Inefficiencies.pdf

Bhavani

T. A.

, & Tendulkar

S. D.

(2001). Determinants of firm-level export performance: A case study of Indian textile garments and apparel industry. The Journal of International Trade & Economic Development, 10(1), 65–92. 10.1080/0963819001001527

Bheda

, Narag

A. S.

, & Singla

M. L.

(2001). Apparel manufacturing productivity in India: Characteristics of the low, medium and high performers. The Journal of The Textile Institute, 92(2), 179–187. 10.1080/00405000108659607

Chapelle

, & Plane

(2005). Technical efficiency measurement within the manufacturing sector in Côte d’Ivoire: A stochastic frontier approach. Journal of Development Studies, 41(7), 1303–1324. 10.1080/00220380500170964

Charnes

, Cooper

W. W.

, & Rhodes

(1978). Measuring the efficiency of decision making units. European Journal of Operations Research, 2(6), 429–444. http://www.utdallas.edu/~ryoung/phdseminar/CCR1978.pdf

Coelli

T. J.

, Rao

P. D.

, O’Donnell

C. J.

, Battese

G. E.

(2005). An introduction to efficiency and productivity analysis (2nd ed.). Springer.

Cooper

D. R.

, Schindler

P. S.

, & Sharma

J. K.

(2012). Business research methods (11th ed.). Tata McGraw Hill Education Private Limited.

D’Souza

(2016). Setting the stage. In Lopez-Acevedo

& Robertson

(Eds.), Stitches to riches? Apparel employment, trade, and economic development in South Asia. Directions in Development. World Bank. 10.1596/978-1-4648-0813-5.

Deshmukh

, & Pyne

P. K.

(2013). Labour productivity and export performance: Firm-level evidence from Indian manufacturing industries since 1991 [Working Paper Series No. 126]. ESCAP. http://www.unescap.org/sites/default/files/AWP%20No.%20126.pdf

10.

Enterprise Surveys. (2016). Data. Enterprise Surveys. The World Bank. http://www.enterprisesurveys.org

11.

Farrell

M. J.

(1957). The measurement of productivity efficiency. Journal of Royal Statistical Society, 120(3), 253–81. http://www.aae.wisc.edu/aae741/Ref/Farrell%201957.pdf

12.

Gambhir

, & Sharma

(2017). Micro enterprises in Indian apparel manufacturing: An efficiency assessment. In Seth

, Singh

& Shukla

(Eds.), Leveraging big data analytics for global business excellence (pp. 134–142). UDH Publishers & Distributors Pvt. Ltd.

13.

Grifell-Tatjé

& Lovell

C. A. K.

(1995). A note on the Malmquist productivity index. Economics Letters, 47(2), 169–175. 10.1016/0165-1765(94)00497-P

14.

Haidar

J. I.

(2012). Trade and productivity: Self-selection or learning-by-exporting in India. Economic Modelling, 29(5), 1766–1773.

15.

Hashim

D. A.

(2005). Post-MFA: Making the textile and garment industry competitive. Economic & Political Weekly, 40(2), 117–127. 10.2307/4416041

16.

Hill

, & Kalirajan

K. P.

(1993). Small enterprise and firm level technical efficiency in the Indonesian garment industry. Applied Economics, 25, 1137–1144. 10.1080/00036849300000174

17.

Joshi

R. N.

, & Singh

S. P.

(2010). Estimation of total factor productivity in the Indian garment industry. Journal of Fashion Marketing and Management, 14(1), 145–160. 10.1108/13612021011025474

18.

Joshi

R. N.

, & Singh

S. P.

(2012). Technical efficiency and its determinants in the Indian garment industry. The Journal of the Textile Institute, 103(3), 231–243. 10.1080/00405000.2011.564797

19.

Kapelko

, & Lansink

A. O.

(2014). Examining the relation between intangible assets and technical efficiency in the international textile and clothing industry. The Journal of the Textile Institute, 105(5), 491–501.

20.

Kumar

(2006). A decomposition of total productivity growth: A regional analysis of Indian industrial manufacturing growth. International Journal of Productivity and Performance Management, 55(3/4), 311–331. 10.1108/17410400610653255

21.

Kundi

, & Sharma

(2015). Efficiency analysis and flexibility: A case study of cement firms in India. Global Journal of Flexible Systems Management, 16(3), 221–234.

22.

Lopez-Acevedo

& Robertson

(Eds.). (2016). Stitches to riches? Apparel employment, trade, and economic development in South Asia. Directions in Development. World Bank. 10.1596/978-1-4648-0813-5.

23.

Mallick

, & Yang

(2013). Productivity performance of export market entry and exit: Evidence from Indian firms. Review of International Economics, 21(4), 809–824.

24.

McDonald

(2009). Using least squares and tobit in second stage DEA efficiency analyses. European Journal of Operational Research, 197(2), 792–798.

25.

Mok

, Yeung

, Han

, & Li

(2010). Export orientation and technical efficiency: Clothing firms in China. Managerial and Decision Economics, 31, 453–463. 10.1002/mde.1500

26.

Ngo

Q. T.

, Tran

Q. V.

, Nguyen

T. D.

, & Nguyen

T. T.

(2020). How heterogeneous are the determinants of total factor productivity in manufacturing sectors? Panel-data evidence from Vietnam. Economies, 8(3), 57. 10.3390/economies8030057

27.

Ramanathan

(Ed.). (2003). An introduction to data envelopment analysis: A tool for performance measurement. Sage Publications.

28.

Ramaswamy

K. V.

, & Gereffi

(1998). India’s apparel sector in the global economy: Catching up or falling behind. Economic & Political Weekly, 33(3), 122–130.

29.

Sarkis

(2007). Preparing your data for DEA. In Modeling data irregularities and structural complexities in data envelopment analysis (pp. 305–320). Springer.

30.

Satpathy

L. D.

, & Mishra

B. R.

(2019). Size-competition-productivity nexus: Evidence from Indian manufacturing firms. South Asia Economic Journal, 20(2), 303–321. 10.1177/1391561419859185

31.

Sharma

, Momaya

, & Manohar

(2010). Assessing the performance of telecommunication industry in India: A data envelopment analysis. Journal of International Business and Economy, 11(2), 29–47. http://www.i-jibe.org/achive/2010fall/2010_11(2)_2.pdf

32.

Van Biesebroeck

(2007). Robustness of productivity estimates. The Journal of Industrial Economics, 55(3), 529–569.

33.

Vixathep

, & Matsunaga

(2012a). Firm efficiency in Cambodia’s garment industry on the eve of the multi-fiber arrangement termination. Asian Economic Journal, 26(4), 359–380. 10.1111/j.1467-8381.2012.02090.x

34.

Vixathep

, & Matsunaga

(2012b). Firm performance in a transitional economy: A case study of Vietnam’s garment industry. Journal of the Asia Pacific Economy, 17(1), 74–93. 10.1080/13547860.2012.640002

35.

Wadud

I. K. M. M.

(2004). Technical efficiency in Australian textile and clothing firms: Evidence from the business longitudinal survey. Australian Economic Papers, 43(3), 357–378. 10.1111/j.1467-8454.2004.00235.x

36.

Waldkirch

(2014). Foreign firms and productivity in developing countries [Working Paper]. Department of Economics, Colby College.

37.

World Bank (2016, April 28). Apparel manufacturing has potential to create 1.2 million new jobs, says World Bank report. Press Release. http://www.worldbank.org/en/news/press-release/2016/04/28/apparel-manufacturing-has-potential-to-create-1point2-million-new-jobs-world-bank-report

38.

World Bank (2017). Country classification by income. Knowledge Base. https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups

39.

World Trade Organisation (WTO). (2017). Clothing exports [Data]. WTO Statistics database.