EXPORTS OF TEXTILE INDUSTRY IN AMRITSAR AND LUDHIANA

Transcription

1 EXPORTS OF TEXTILE INDUSTRY IN AMRITSAR AND LUDHIANA This chapter is divided into two sections. The first section deals with textile exports of Amritsar and Ludhiana vis-a-vis Punjab and the second section empirically deals with determinants of exports of textile firms in Amritsar and Ludhiana in wake of ATC phase out. The Indian textile industry has come a long way and achieved unique status on the global textile map and created a niche position for itself, in the process resulting into enhanced FDI inflows, increased joint ventures, inflow of superior technical know-how, etc. As a provider of jobs, it is second only to the railways. Known globally for its skill and craftsmanship, the Indian textile industry accounts for more than 5 per cent GDP, 20 per cent of manufacturing value added and one third of total export earnings. (Landes et.al, 2005). The share of Indian textile in world falls within a range of 2.5 to 3 percent. China is one of the strong competitors in the international market. China produces raw cotton as many as 4 to 5 times that of India, woollen production (5 to 6 times); spun yarn (3 times) and woven fabrics (1-4 times) exceeded our national production. Its garment industry is 3 times bigger and its textile industry is 9.5 times bigger than ours (Singh & Kundu, 2005). The Multifibre agreement (MFA) has governed international trade in textiles and clothing since The MFA enabled developed nations, mainly USA, European Union and Canada to restrict imports from developing countries through a system of quotas. The Agreement on Textiles and Clothing (ATC) to abolish MFA quotas marked a significant turnaround in the global textile trade. Under ATC, there was progressive phase out of import quotas which were established under MFA and the textiles and clothing was to be integrated into the multilateral trading system before January ATC is a transitory regime between MFA and the integration of trading in textiles and clothing in the multilateral trading system. The ATC provided for a stage wise integration process to be completed within a period of ten years ( ), divided into

2 four stages starting with the implementation of the agreement in The product groups from which products were to be integrated at each stage of integration included (i) tops and yarns (ii) fabrics (iii) made-up textile products (iv) clothing.(exim bank :Research Brief,2005) TABLE 7.1 FOUR SHORT STEPS TOWARDS FREEING TEXTILE TRADE Step Step-1 1 Jan 1995 (to 31 Dec 1997) Step-2 1 Jan 1998 (to 31 Dec 2001) Step-3 1 Jan 2002 (to 31 Dec 2004) Step-4 1 Jan 2005 Full integration into GATT Percent age of products to be brought under GATT (including removal of any quotas) 16 per cent (minimum, taking 1990 imports as base) How fast remaining quotas should open up, if 1994 rate was 6 percent 6.96 percent per year 17 per cent 8.7 percent per year 18 per cent percent per year 49 per cent (maximum) No quotas left (and final elimination of quotas) Source : WTO: Trading into Future,1997 (Cotton textile industry in India- Banik and Bandopadhyay, 2000) SECTION I The most important component of textile industry in Punjab is cotton and woolen textiles. The cotton textiles industry is complex and varied in its structures, with the hand spun, hand woven sector at one hand and of the spectrum and sophisticated capital intensive high operation on the other. In textile industry, handlooms, power looms and automatic machines are used simultaneously by different units in weaving process and it will be of interest to access the role of technological progress in the textiles goods industry in Punjab which is one of the most important manufacturing centers of these goods in India. In Punjab, Amritsar is famous for the manufacturing of fine pashmina shawls, thick serge, silk goods and carpets. Besides the pashmina work, Amritsar is also engaged in manufacturing of silk goods. Carpet weaving is another industry, which is very popular 129

3 in Amritsar. Industrialization is occurring at a fast pace in the city. The 1950s saw a huge expansion in the industrial sector of Amritsar with many new industries coming up and old ones expanding at a fast rate. Ludhiana textile has been dressing up India for a long time. The Ludhiana textile industry comprises of exporters, brand producers and high scale manufacturers. It is a base for all big brands - Madame, Monte Carlo, Casablanca, Trident, Rage, Duke, Sportking and Godani. The industry employs more than five hundred thousand skilled workers. The textile industry of Ludhiana, thus, provides employment to many people. It is also the backbone of Ludhiana's economy (District Industries Centre, Ludhiana) In terms of Human Development Index (HDI), Ludhiana ranks first whereas Amritsar ranks nine in Punjab with HDI.761 and.700 respectively. (data.undp.org) Table 7.2 shows distribution of SSI and Large and Medium industry in Amritsar and Ludhiana. It can be seen that number of industry in Ludhiana (both SSI and L&M) is far more than Amritsar though their corresponding number has decreased in 2009 viz-aviz Employment has marginally increased in 2009 as compared to 2000 in SSI of both Amritsar and Ludhiana. In case of L&M industry, employment has actually decreased in case of Amritsar. As far as investment and production are concerned, they have decreased in 2009 as compared to 2000 both for Amritsar and Ludhiana. Table 7.3(a) shows that percentage share of Ludhiana exports in Punjab is much higher than percentage share of Amritsar exports. From period to , share of Ludhiana exports in Punjab was the highest in i.e percent whereas share of Amritsar exports was the highest in i.e percent. Since 70 percent of exports in Punjab are from Ludhiana, this shows export concentration which is a structural characteristic of underdeveloped economies. Year to year growth rates of both districts show that Amritsar exports were the highest in i.e percent and Ludhiana exports were highest in i.e percent. This may be because the merchandise exports had accelerated since 2003, largely due to rise in oil and commodity prices. The trends of total exports in the sample districts have been shown in graph Fig

4 TABLE 7.2 DISTRIBUTION OF SMALL SCALE INDUSTRIES (SSI) AND LARGE AND MEDIUM INDUSTRIES (L&M) IN AMRITSAR AND LUDHIANA SSI Year District SSI (No.) Employment (No.) Investment (Rs. Cr.) Production (Rs. Cr.) 2000 Amritsar Ludhiana Amritsar Ludhiana L&M Year District L&M (No.) Employment (No.) Investment (Rs. Cr.) Production (Rs. Cr.) 2000 Amritsar Ludhiana Amritsar Ludhiana Source: Directorate of industries, Punjab, Chandigarh, District Industries Centers, Amritsar and Ludhiana. 131

5 TABLE 7.3 (a) DISTRICT WISE EXPORTS FROM AMRITSAR AND LUDHIANA Year Amritsar Ludhiana Punjab % Share of Amritsar exports in Punjab (Amount in Rs. Crore) % Share of Ludhiana exports in Punjab (8.75) (-18.22) (32.22) (1.52) (-6.34) (38.43) (18.91) (-28.33) (15.82) (28.96) (7.29) (49.62) (17.29) (13.71) (-18.48) (10.19) (-2.20) (13.75) (44.75) (55.17) (.740) (27.08) (25.09) (-12.93) (22.82) (9.81) (15.48) (-13.69) (11.94) (-1.17) (9.78) (59.11) (27.37) (-11.40) (22.00) (22.18) (-9.21) (29.67) (15.00) Source: Directorate of industries, Punjab, Chandigarh, District Industries Centers, Amritsar and Ludhiana. Note: Figures in parentheses are yearly growth rates

6 Table 7.3(b) shows that the compound growth rate of Amritsar exports in Post-ATC or after quota abolition was more i.e percent whereas in pre-atc period, it was 6.45 percent. t-value is also significant meaning thereby that quota abolition has helped to increase exports in Amritsar. In case of Ludhiana, the growth in post-atc is only 8.72 percent whereas in pre-atc, it was percent, which means that quota removal has decreased the growth rate of exports. However, the overall growth rate for the whole period is 9.69 percent and percent for Amritsar and Ludhiana respectively. Punjab is heavily dependent on few products where it enjoys some degree of comparative advantage. Its range of exports was usually limited to a few labour intensive industries, mostly in clothing. The vulnerability of Punjab economy can also be judged from the fact that out of its total exports, major portion was of textiles. Table 7.4 depicts textile exports of both Amritsar and Ludhiana for to It can be seen that both these districts show negative growth rates in and This may be due to biased policy of the Government. Most industry in Punjab is power and labour intensive. In peak summers, there are long power cuts for industry, as the available power is diverted to agricultural sector, especially for paddy cultivation. This increases the cost of production, as own power generation is very expensive for industrial units. As far as percentage share is concerned, it can be seen that share of Ludhiana textile export in total export of Punjab was around 50-60%. The compound growth rate of textile exports in Amritsar was negative, but for Ludhiana textile exports, it was positive. However, their t-values were not significant. 133

7 FIG. 7.1 EXPORTS FROM AMRITSAR, LUDHIANA and PUNJAB Source: Data from Directorate of industries, Punjab, Chandigarh, District Industries Centers, Amritsar and Ludhiana. TABLE 7.3 (b) GROWTH RATES (COMPOUND) OF EXPORTS Period Amritsar Ludhiana Punjab ( to ) t-value Pre ATC( to ) 7.55* * * t-value Post ATC( to ) 3.19* * * t-value 8.96* Note: 1) * Significant at 1 percent level 2) ** Significant at 5 percent level 2.83** 4.02* 134

8 TABLE 7.4 DISTRICT WISE TEXTILE EXPORTS FROM AMRITSAR AND LUDHIANA (Amount in Rs. Crore) Year Amritsar Ludhiana Punjab % Share of Amritsar in Punjab % Share of Ludhiana in Punjab (9.60) (42.26) (.810) (-59.00) (-14.51) (-8.02) (-35.00) (-17.81) (8.00) (165.87) (55.17) (27.84) (7.71) (8.86) (8.72) CAGR t-value ** Source: Directorate of Industries, Pun jab, Chandigarh District Industries Centre Amritsar and Ludhiana Note: 1) Figures in parentheses are yearly growth rates 2) ** Significant at 5 percent level The trends of textile exports in Amritsar and Ludhiana are shown in Fig

9 FIG. 7.2 TEXTILE EXPORTS FROM AMRITSAR, LUDHIANA and PUNJAB Source: Data from Directorate of industries, Punjab, Chandigarh, District Industries Centers, Amritsar and Ludhiana. Table 7.5 shows that the share of Amritsar textile exports in its total exports ranged from 5%-14% whereas Ludhiana textile exports in its total exports ranged from 30%-46% in the period from to This is true as dominating product in Ludhiana exports is textiles. 136

10 TABLE 7.5 SHARE OF TEXTILE EXPORTS IN TOTAL EXPORTS OF AMRITSAR AND LUDHIANA Year Textile export Amritsar Total exports Amritsar Textile export Ludhiana Total export Ludhiana Share of Amritsar textile exports in Amritsar total exports Share of Ludhiana textile exports in Ludhiana total exports (9.60) (15.81) (42.26) (27.08) (-59.00) (28.96) (-14.51) (25.09) (-35.00) (7.29) (-17.81) (-12.93) (165.87) (49.62) (55.17) (22.82) (7.71) (17.29) (8.86) (9.81) CAGR t-value Source: Directorate of Industries, Punjab, Chandigarh District Industries Centre- Amritsar and Ludhiana Note: 1) Figures in parentheses are yearly growth rates 137

11 SECTION II DETERMINANTS OF EXPORTS OF TEXTILE INDUSTRY IN AMRITSAR AND LUDHIANA IN WAKE OF PHASE OUT OF ATC This section analyses the determinants of textile exports during of the cross sectional sample units located in Amritsar and Ludhiana. In this study, Panel Regression models have been estimated. An advantage of using Panel data regression is that more observations on explanatory variables are available. This has the effect of helping to overcome the inherent multicollinearity, which probably exists between the independent variables in OLS estimations. The Panel data (also called longitudinal data) refers to the data for n different entities observed at T different time periods. Here the entities are the textile firms 45 in case of Amritsar and 55 in case of Ludhiana over the period i.e. T=9. So there are (45x9=405 observations) in case of Amritsar and (55x9=495 observations) for Ludhiana and (45+55)x9=900 observations for both Amritsar and Ludhiana. The analysis of determinants of textile exports have been examined by estimating the following models Model I ln Tex = a o +a 1 ln S+a 2 lnfi + a 3 lnw+a 4 lna+a 5 lnwd + a 6 lnrp + a 7 ln REER + ATC + µ 1 Model II ln Tex = a o +a 1 ln O+a 2 lnfi + a 3 lnw+a 4 lna+a 5 lnwd + a 6 lnrp + a 7 ln REER + ATC + µ 1 138

12 TABLE 7.6 DESCRIPTION OF VARIABLES USED IN REGRESSION ANALYSIS Variable Description Hypothesized relationship with textile exports S Sales + O Output + FI Fixed Investment + W Wages + A Age + WD World Demand + RP Relative Price - REER Real Effective Exchange Rate - ATC L t Agreement on Textiles and Clothing Location Time ATC, L are dummy variables. ATC ( ) = 0 and 1 other wise. Location has been used as dummy variable in pooled panel data analysis (0 for Amritsar and 1 otherwise). The estimated coefficients represent relative elasticities. It is hypothesized that a 6, a 7 < 0, while a 1, a 2, a 3, a 4, a 5 >0 139

13 Panel Regression equations have been estimated for Amritsar firms and Ludhiana firms separately and in the end, the data is pooled up and again with the help of panel regression, it is checked as to which variables affect exports significantly. Measurement of variables a) Sales (S): Sales play a significant role in determining export performance (Teresa,2006) The sales figures have been deflated with wholesale price index (Base ). Sales have been used as a proxy for size. Higher sales indicate larger size which implies economies of scale and higher potential for exports (Lefebvre and Lefebvre, 2001) (Chandler, 1990). Larger firms are generally regarded as more capable of bearing the large investments and high risks associated with exporting. Several studies support his view empirically,i.e. (Dholankia and Kapur,1999(Aitken,Hanson and Harrison,1997,Roberts and Tybout,1997,Hirsch and Adar,1974, Bernard and Wagner,1996 and Lall and Kumar,1981).The expected sign of its coefficient is positive. b) Output (O): The output or total production of a firm has considerable impact on exports. It has also been deflated by WPI with ( ) as base. Its coefficient is also expected to be positively related with exports. c) Fixed Investment (FI): In this study, only small scale industries are taken (that is, whose investment in plant and machinery is less than Rs 5 cr). The data of FI has been deflated with WPI(Wholesale Price Index). Its coefficient is expected to be positively related with quantity of exports. d) Wages (W): Wages as a variable has been used as a proxy for quality of labour force, assuming higher the wage, more educated and skilled is the workforce. The data of wages has been deflated using Consumer Price Index (CPI) Base ( ). This should influence positively exports of a firm. e) Age (A): The age of the firm is a debated factor. Some studies have failed to come up with a correlation between age and export performance [Ong and Pearson (1982)], others have found positive relationship [Abbas and Swierz (1991)], and others have confirmed a negative one [Ursic and Czinkota (1984)]. According to Nassimbeni (2001), this is the result of two contrasting effects. On one side, a firm s age goes hand in hand with operational and commercial 140

14 experience and solidity and, therefore, with a more adequate approach to export activity. On the other, young firms are usually leaner and more receptive to changing perspectives. The coefficient of age of the firm is expected to bear positive sign (Papadogonas et.al., 2007). f) World Demand (WD): The exports of a country are also influenced by world income. World exports have been taken as a proxy for world demand. World exports have been taken from and this was given in US Dollars. Exchange rate for said period was taken from India Infrastructure Data base (2005). Data on world exports were converted into Indian rupees. Sign of coefficient of this variable in the regression equation is expected to be positive. g) Relative price (RP): Relative price of textiles has been calculated by dividing manufactured unit value index by wholesale or domestic price index and multiplying this with 100. Manufactured unit value index was taken form Handbook of Statistics and Indian Economy and Domestic price Index was taken from Economic survey ( ). These indices were prepared with base ( ). Its expected sign in regression equation is negative.( Nanda,1998) h) Real Effective Exchange Rate (REER): The real effective exchange rate has been taken from Economic Survey ( ).A depreciation of REER increases the relative profitability of producing tradable goods. Therefore, the expected sign of its coefficient is negative i) ATC: Agreement on Textiles and Clothing has been taken as a dummy variable. From , ATC is O and I otherwise. Its effect was calculated for Amritsar firms and Ludhiana firms individually and its effect was also seen when they were pooled up. j) Location (L): This is another dummy variable which has been used only in pooled data. For Amritsar firms, its value is O and I otherwise. The results are calculated and interpreted subsequently. 141

15 TABLE 7.7 PANEL REGRESSION ESTIMATES OF DETERMINANTS OF AMRITSAR (SAMPLE FIRMS) Variable Model-I Model-II Constant (1. 599)*** (2.024)* WD : World Demand.230 (0. 732) NS.105 (.314) NS S : Sales (29.291)* FI : Fixed Investment (-.555) NS (-.228) NS A : Age.017 (.337) NS.017 (.309) NS REER :Real Effective Ex.Rate (-1.171) NS (-1.624)*** RP: Relative Price (-2.774)** (-2.861)** W : Wage (-5.092)* (-4.188)* O :Output (26.753)* ATC.101 (1.085) NS.131 (1.328) NS Number of observations 405 R DW Note: 1) Figures in parentheses are t-values 2) * Significant at 1 percent level 3) ** Significant at 5 percent level 4) *** Significant at 10 percent level 5) NS Not significant. It can be seen from table 7.7 that in both models, elasticity of sales and output are significant for Amritsar Textile Firms. The coefficient of both sales and output shows that 10 percent increase in sales and output would lead to 12 percent increase in exports. All the coefficients bear hypothesized signs. The coefficient of fixed investment is negative but not significant. However, coefficient of wage is negative but significant. This can be attributed to the fact that exporters in Amritsar are mainly labour intensive, low technology manufacturing firms, which actually do not need skilled or highly 142

16 educated labour force. This confirms the results of Papadogonas et. al. (2007) and Liu and Shu (2001). The coefficient of RP is significant at 5 percent level. ATC does not have significant impact on exports of textiles in Amritsar for the time period as indicated by the non significant coefficient of the differential coefficient of ATC in case of both the models. This indicates that quota phase out did not have any impact on the textile exports of Amritsar firms. The variables included in Model-I and Model-II explain 89 and 88 percent of the variation respectively in the textile exports of the sample firms in Amritsar. TABLE 7.8 PANEL REGRESSION ESTIMATES OF DETERMINANTS OF LUDHIANA (SAMPLE FIRMS) Variable Model-I Model-II Constant (1.600)*** (1.458)*** WD : World Demand.436 (1.338) NS.541 (1.590)*** S : Sales.859 (23.819)* FI : Fixed Investment.013 (.144) NS.106 (1.118) NS A : Age.186 (2.792)**.185 (2.637)** REER :Real Effective Ex.Rate (-1.079) NS (-1.046) NS RP: Relative Price (-2.363)** (-2.105)** W : Wage.378 (4.765)*.495 (6.096)* O :Output.781 (21.881)* ATC.087 (.898) NS.071 (.703) NS Number of observations 495 R DW Note: 1) Figures in parentheses are t-values 2) * Significant at 1 percent level 3) ** Significant at 5 percent level 4) *** Significant at 10 percent level 5) NS Not significant. Table 7.8 shows results of the Panel Regression estimates of exports in case of textile units surveyed in Ludhiana In Model-I, sales and wage are both significant and bear expected signs. The coefficient of wage indicates that firms in Ludhiana hire more skilled and technical labour as compared to Amritsar. Age turned out to be a significant 143

17 factor in Ludhiana firms. It shows that experience affects export propensity positively. The differential coefficient of ATC in case of Ludhiana sample units implies that removal of quota restrictions has not put any significant impact on its textile exports as shown by non significant coefficient. The model is able to explain 78 percent of the variation. The results of Model-II do not show much difference as far as the significance of the variables is concerned. TABLE 7.9 PANEL REGRESSION ESTIMATES OF DETERMINANTS OF AMRITSAR AND LUDHIANA (POOLED) Variable Model-I Model-II Constant (2.313)*** (2.343)*** WD : World Demand.404 (1.700)***.437 (1.736)*** S : Sales (37.724)** FI : Fixed Investment.153 (3.946)*.225 (5.550)* A : Age.100 (2.328)**.100 (2.196)** REER :Real Effective Ex.Rate (-1.693)*** (-1.833)*** RP: Relative Price (-3.696)* (-3.477)* W : Wage (-.775) NS.030 (.623) NS O :Output.937 (34.186)*** ATC.107 (1.515)***.109 (1.455)* L : Location (-6.670)* (-7.364)* Number of observations 900 R DW Note: 1) Figures in parentheses are t-values 2) * Significant at 1 percent level 3) ** Significant at 5 percent level 4) *** Significant at 10 percent level 5) NS Not significant. In Table 7.9, Model-I shows that sales are significant at 1 percent level. The coefficient of wage is negative but not significant. Fixed investment shows positive and significant relationship with exports. Coefficient of RP bears negative sign as 144

18 hypothesized and is significant. It means if elasticity of relative export price to domestic price decreases, there is increase in exports. Age is also significant at 5 percent level. Location was used as dummy to capture its effect on the firms in Amritsar and Ludhiana. Its coefficient is negative and significant. It indicated that firms located in Amritsar and Ludhiana were significantly different from each other. This was because the exports of Ludhiana during the said time period had suffered a lot and its textile industry had become stagnant in its growth over the past few years. It needed technology up gradation and a push from the government for achieving new milestones. It faced an unusual situation in 2010 when their US based buyers cancelled orders, alleging that these units employed children. Around 20 per cent of the orders got cancelled. It had over 170 licenses of "Woolmark" at one point of time. Woolmark, is the standardization given for purity of wool in the product. The number came down to just 15. The decreasing number reflected how the demand of pure wool had reduced over the years (NITMA, 2008). The coefficient of ATC was significant at 10 percent level indicating that quota phase out had a positive and significant impact on exports of textile units of both Amritsar and Ludhiana when the sample units were pooled up. The DW statistic in all these cases also fell between lower and upper critical values at 1 percent level. Therefore, no conclusive inference could be drawn with respect to serial correlation of the error term. The empirical results show that the exports of textiles in Punjab in post-atc era have not been commensurate with its expectations by any reasonable yardstick. Main factors responsible for this are that Punjab s textile has not been able to rise up to the occasion and its poor capabilities and resources have inhibited it from moving beyond the level of other competing developing countries.also, the vision of Punjab textiles and clothing was very myopic about the post-quota gains whereas they expected the gains to accrue to them even as they did very little to ensure them. In order to ensure deserving success in future, Government support is badly required by small scale sectors (Rameshan, 2005). Also, textile industry needs faster product innovation, quick response to changes in consumer preferences and should have the ability to move up in value chain by building brand names and acquire channels of distribution so as to outweigh the advantages of competitors in the long run. *** ** 145