Exchange rate movements and export market dynamics: evidence from China

This paper highlights the relationship between foreign exchange rate fluctuations and firms’ export market dynamics using a Chinese firm-level production data and a firm-level trade data over the period of 2000–2006. The author adopts a discrete-time survival model in his empirical investigation and further executes several extensions and robustness checks to the baseline results. The main results of the paper can be summarized as follows: First, an exchange rate appreciation increases the likelihood of export market exit, reduces the capability of export market survival and decreases the probability of export market entry. Second, high productivity firms are less likely to exit from export markets and more likely to enter and survive in export markets in the period of exchange rate appreciation. Third, exchange rate appreciation decreases the likelihood of export market entering and increases the likelihood of export market exiting more for private-owned firms, young firms and non-eastern firms. Finally, other sources of heterogeneity, such as extensive margins, import demand elasticity, different destinations, U.S. dollar peg, and the liberalization of trading rights also matter regarding the effect of exchange rate changes. JEL F14 F31 F32 F41


Introduction
It is well known that exchange rate is a significant institutional factor affecting aggregate trade and individual exporting behavior. In reality, facing exchange rate movements, exporters not only adjust their export volume and export price, but also by switch their products mix or even export market dynamics (entry or exit). Understanding the influence of exchange rate fluctuations on the international trade is of special interests to both researchers and policy makers, especially in the wave of global imbalance. A large and growing number of studies have shed light on the effect of foreign exchange rate fluctuations on export performance including export volume (the exchange rate elasticity of export quantity) and export price (exchange rate pass-through, ERPT) (Shambaugh, 2008;). 1 Another strand of literature highlights the relationship between exchange rate movements and extensive or intensive margin of trade on the product side (Bernard and Jensen, 2004a;Beggs et al., 2009). In the vast relevant literature there are even some evidences that link the effect of exchange rate movements to firm-level characteristics trigged by the growing studies of firm's heterogeneity (Das et al., 2007;Bernard et al., 2011) and the better availability of firm-level data, such as Berman et al. (2012) and Li et al. (2015). But very little attention is paid to the reaction of exporters to foreign exchange rate movements in terms of export market dynamics.
China's foreign trade and trade surplus experience fast growth since the reform and opening-up policy began, meanwhile Chinese currency (RMB) has been appreciating in recent years according to the statistics from Bank of International Settlements (see Picture 1). China's ballooning current account surplus and rapid accumulation of international reserves has been under the limelight for a long time. Many concern whether Chinese authorities are heavily managing their currency and contributing to global imbalances. Many also question whether faster currency appreciation would reduce China's trade surplus tremendously. However, relevant studies are still inconclusive owing to different empirical methods and data coverage. The main results of the paper can be summarized as follows: First, an exchange rate appreciation increases the likelihood of export market exit, reduce the capability of export market survival and decreases the probability of export market entry. Second, high productivity firms are less likely to exit from export markets and more likely to enter and survive in export markets in the period of exchange rate appreciation. Third, exchange rate appreciation decreases the likelihood of export market entering and increase the likelihood of export market exiting more for private-owned firms, young firms and non-eastern firms. Finally, other sources of heterogeneity, such as extensive margins, import demand elasticity, different destinations, U.S. dollar peg, and the liberalization of trading rights also matter, but our results are robust to alternative redefinition of dynamics, alternative productivity measures and alternative estimation approaches.
We may contribute to relevant literature in three aspects. First, unlike the literature using aggregate data, we use a longitudinal firm-level data merged from a production data and a trade data, which allows us to calculate destination specific exchange rate movements and control more firm-level factors affecting firms' export dynamics. Second, we relate the effects of exchange rate movements on export market dynamics to several firm characteristics, which allows us to gain a better understanding on how exchange rate movements affect export market dynamics and more policy implications. Finally, we are different from previous studies by adopting a discrete time survival model which is a natural framework to address the question of success and failure in export markets with our data.
Our paper most relates to a small but growing number of studies linking exchange rate movements to export market dynamics. Beggset al. (2009) show that the impact of real exchange rate changes on firm survival is far larger than the effect of CUSTA tariff reduction. Tang and Zhang (2012) find that a significant impact of real appreciation of the renminbi on the extensive margins of Chinese exporters. Goerg and Spaliara (2013) find a positive relationship between exchange rate and the hazard of exit in their paper studying the effect of financial pressure on export market exit.
Our research also makes a contribution to recent studies examining the effect of exchange rate on export volume and export price. First, a number of papers investigate the relationship between exchange rate devaluation and export growth, most of them find that large depreciations of the real exchange rate were an important determinant of export surges (Fang et al., 2006;Bernard and Jensen, 2004;Freund and Pierola, 2013;Haddad and Pancaro, 2010). Second, with respect to export price, many studies find that exchange rate fluctuations have small effects on the prices of internationally traded goods, which is considered as the exchange rate disconnect puzzle (Goldberg and Knetter, 1997;Goldberg, 2005, 2008). Possible explanations for incomplete pass through include short-run nominal rigidities (Engel, 2003;Gopinath and Itskhoki, 2010;Gopinath et al., 2010;Gopinath and Rigobon, 2008), pricing-to-market strategies (Atkeson and Burstein, 2008;Knetter, 1989Knetter, , 1993, or local distribution costs (Burstein et al., 2003;Corsetti and Dedola, 2005). Furthermore, some studies shed light on the investigation of the heterogeneous pricing response of exporters to exchange rate changes owing to the increasing availability of firm-level trade data. Amiti et al. (2014) find that exporters with high market shares have a lower exchange rate pass-through. Berman et al. (2012) show that the export prices of highly productive French exporters are more sensitive to real exchange rate changes. Chatterjee et al. (2013) find pricing-to-market is stronger for the products the firm is most efficient at producing.
Our paper also adds to the literature looking at China's exchange rate movements and their exports.
The classic theory of exchange rate states that given that Marshall-Lerner condition satisfies, the 4 appreciation of foreign exchange rates causes an adverse affect on trade surplus. many studies consequently exam whether Marshall-Lerner condition satisfies in China, some of them argue that Marshall-Lerner condition satisfies in China (Liu et al., 2010), while the others hold opposite view (Han, 2010;OECD, 2011b). More studies pay attention to the exchange rate pass through of RMB.
Ca ' Zorzi et al. (2007) show that a 1% change in the exchange rate will lead to a change of 0.08% in China's consumer prices after 4 quarters, and 0.77% in 8 quarters. Xiang (2006) suggest that if the exchange rate appreciates by 1%, producer and consumer prices will decline by 0.2% and 0.1% respectively. Ahmed (2009) find the exchange rate elasticity of RMB is above unity. Li et al. (2015) find almost complete exchange rate pass through into RMB export price for Chinese exporters.
The remainder of the paper is organized as follows. Section 2 introduces the dynamics of Chinese exporting firms. Section 3 describes the datasets we employed in this paper. Section 4 analyzes the baseline empirical results. Section 5 demonstrates some extensions and robustness checks. The last section concludes.

Dynamics of Chinese exporting firms
Exporting activities incur more risks and costs than domestic businesses due to institutional differences, complicate transaction procedures and market fluctuations. This section exam the dynamics of Chinese firms at foreign markets using the filtered production data from 2000 to 2006. We use the information on the yearly export delivery value to identify whether a firm enters into or exits from foreign markets. The appearance of a positive value indicates an entry into foreign markets, and the disappearance of a positive export delivery value indicates an exit from foreign markets. If a firm re-enter into the export market after exits, we treat such a firm as a new entry firm at that year. 2 Table 1    First, the surviving ability of Chinese firms in foreign markets is weaker than domestic firms, about 42% of all firms can survive for 7 years, but only 29% of the firms who export in 2000 succeed to export until 2006. Second, the longer firms are active in foreign markets, the stronger they become. On the one hand, exporting firms are more likely to exit from foreign markets in the beginning years, 25% of exporting firms exit from foreign markets after 1 year, but this number is 6 9% after 6 years. On the other hand, firms who succeed to export for 5 years gain a better performance than firms being exporters for 4 years in terms of export value, sales and employee number. Finally, when comparing exporting firms with all firms (exporters and domestic firms), we find that exporting firms are better in all aspects, which can likely be explained by the self-selection effect and the positive learning externalities because this causality can run in both directions. What we find here serves as an evidence verifying the importance of firm heterogeneity in international trade. Table 2 displays the firm entry and exit rates at foreign markets over the period 2000-2006. We define entrants in year t as firms that are absent in t-1, but appeared in t. We define exiters in year t as firms that are active in t-1 but absent in t. The exit and entry rates are calculated as the share of entering firms and exiting firms in total number over the exporting firms in that year. Note: Export value of exit firms refers to the observations of last year, and the export value is mean value in 1000 RMB.
As indicated in Table 2, the exit and entry of exporting firms in foreign markets are turbulent, the annual turnover rate fluctuates between 46% and 60% over the period of 2000-2006. 3 The number of entrants into foreign markets account on average for 25% of the total number of exporting firms each year, while an average of 28% of the exporting firms exit from foreign markets each year.
3 Firm turnover rate is the sum of entry rate and exit rate.
These figures are very close to the turnover rate of Columbian firms (Eaton et al, 2008), whereas they are much higher than those of many other countries (e.g. Bartelsman et al., 2013). The high churning rate of exporting firms likely reflects their more intense dynamics on exporting markets.

Data
To investigate the impact of exchange rate fluctuations on firm's exports, we employ two Chinese datasets and in our paper. One is the production data from Annual Surveys of Industrial Production and 301,961 firms (in 2006). The dataset not only provides some basic information, such as name, address, age, ownership, but also financial information, such as output, wage, employment added-value, export delivery value, profit and fixed-assets.
The firm-level dataset contains much noisy information. We filter the data by following steps. (1) liquid assets are greater than total assets; (2) total fixed assets are greater than total assets; (3) the net value of fixed assets is greater than total assets; (4) the firm's identification number is missing.
The other dataset we use is the product-level trade data from the China's General Administration of Customs, which covers all exporters and importers from 2000 to 2006. It records a variety of information for each trading firm's product list, including trading price, quantity and unit value at the HS 8-digit level.
We merge the above two databases according to the contact information of firms, because there is no consistent coding system of firm identity between these two databases. Following Yu (2014), we go through two steps to match these two datasets. First, the vast majority of firms are matched by company names exactly. Second, we adopt telephone number and zip code to identify firms as a supplement. Table 3 describes the matched dataset.  (5) reports the number of firms covered in the firm-level production data set compiled by China's National Bureau of Statistics without any filter and cleaning. Column (6) presents number of firms covered in the merged date using the trade data set and the raw production data set.

Specification
To evaluate the effects of foreign exchange rate movements on export market dynamics, we use a complementary log-log model (cloglog), which is a discrete-time version of the Cox proportional hazard model. 5 We prefer this model because the annual nature of the data causes right-censoring: , shows the interval hazard for the period between the beginning and the end of the th j year after the first appearance of the firm. This hazard rate, which is the rate at which firms fail at time t given that they have survived in 1 t  , takes the following form: Where the identification of  parameters is our primary interest, which exhibit the effect of the explanatory variables on the hazard rate.
We begin with a benchmark model to connect the firms' probability of entry and exit the export markets to exchange rate movements and other control variables.
The sign and significance of 1  shows the direction and importance of exchange rate movements on export market dynamics. Positive estimates normally suggest that the larger values of the explanatory variables increase the hazard, or equivalently, decrease the probability of survival. But this is not the case for RMB exchange rate movements, positive estimates of RMB appreciation suggest a negative impact of RMB appreciation on the hazard, since a decrease of RMB exchange rate implies an appreciation of RMB against foreign currency.
We define entrants in year t as those firms don't export in year 1 t  but in year t and exiters in year t as those firms that export to foreign markets in year 1 t  but not in year t . We treat firms as new entrants if they reenter into export markets after exit.
With respect to the independent variables, REER t is the real effective exchange rate of RMB against country in year t . 1 it Z  denotes the vector of control variables including firm productivity, firm size, firm age. In order to deal with the lagged effect of productivity and the possible concern of endogeneity, we include their one year lagged term of firm productivity. Firm size is measured by the number of employees (Liu and zhang, 2008). Firm age is measured by the difference between firm establishing year and the current year. Moreover, we include a full set of time, ownership, industry and regional dummies in order to control many firm-specific and year-specific fixed factors. 6 Firms are heterogeneous in their productivity and other firm characteristics, therefore their reactions to exchange rate movements may also be heterogeneous. We modify the benchmark model to interact exchange rate movements with firm productivity and other control variables in order to assess how the effect of exchange rate movements relates to firm heterogeneities.
We report the basic statistical information of key variables in Table 4.

Measurement of TFP
There are several methods for productivity estimation including Solow's residual method, data envelopement analysis (DEA) method, Olley-Pakes (OP;1996) method, and Levinsohn-Petrin (LP; method. Solow's residual method is most used for its simplicity, but it generates simultaneity bias and selectivity bias. Olley and Pakes (1996) proposed a semi-parametric estimator to reduce simultaneity bias, which has become the most popular method for estimating firm productivity.
In this section, we also adopt the OP method to estimate firm productivity using added value as the dependent variable. We use fixed assets and the number of employees as measures of the explanatory variables capital and labor. We utilize the perpetual inventory method to calculate capital stocks assuming a 15% depreciation rate. 8 All variables are deflated by appropriate price indices. Our paper executes Solow residual (OLS) and OP estimations for comparison. The coefficients of capital and labor are listed in Table 5. Note: t-values is in parentheses. Significant at * 10%, ** 5% and*** 1%. Olley and Pakes (1996) state that simultaneity bias and selectivity bias generated by OLS estimation cause an upward bias for the labor coefficient and a downward bias for the capital coefficient. As shown in Table 3, the capital coefficient is indeed higher for OP than for OLS, while on the contrary, the labor coefficient is lower for OP than for OLS. The estimation results, thus, conform with the conclusion of Olley and Pakes (1996), which makes us confident that the risk of biased productivity estimates is considerably reduced by the use of OP estimation.
8 Some papers adopt other lower depreciation rates, such as 10% or 5%. The choice of different depreciation rates does not affect our qualitative results.

Exchange rate movements and export market exit
The connection between exchange rate movements and the export market exit is explored in this section. Table 6 reports the estimation results by adopting probit regression. Column 1 presents the estimation results without considering the firm-industry-year triplet specific fixed effects, column 2 includes these specific fixed effects, and column 3-6 interact the exchange rate movements with firm heterogeneities including productivity, age and dummies.
As noted in Table 6, we find that the coefficients of exchange rate movements in all specifications are negative and significant at the 1% level, implying that exchange rate changes play a positive effect on the likelihood of firms exiting from the export markets. To be specific, an appreciation in RMB by 10% (a decrease of RMB exchange rate) leads to an increase in the hazard of export market exit by about 60%. Moreover, as indicated in the table, the estimated coefficients of the interaction term between exchange rate and firm productivity is found to be negative and significant, which suggests a weaker effect of exchange rate appreciation on export market exit to high productivity firms, namely, high productivity firms are less likely to exit from export markets in the period of exchange rate appreciation.
lnreer -4.085*** -3.932*** -6.258*** -6.262*** -6.462*** -6.562*** (  Note: Robust t-statistics in parentheses. *** p<0.01, ** p<0.05, * p<0.1 As indicated by the interactions between exchange rate movements and firm-specific effects, we find that private firms gain stronger response to the exchange rate volatility, while the responses of younger firms and eastern firms are weaker than their counterparts. This result shows that facing the exchange rate appreciation (1) private-owned enterprises are more likely to exit from export markets, (2) younger exporters are more likely to exit from export markets, (3) firms from eastern China enjoy lower propensity of failure in export markets.
As for other controls, we find a negative relationship between the probability of exiting and firm age and firm size, although their effects are of less importance for the export market exit. Table 7 shows the estimation results of the relationship between the exchange rate fluctuations and the export market entry by using probit regression. The first column presents the estimation results without considering the firm-industry-year triplet specific fixed effects, while column 2 includes them, and column 3-6 relates the impact of exchange rate movements to firm heterogeneity by introducing the interaction terms.

Exchange rate movements and export market entry
From Table 7, we observe that the variable of RMB exchange rate movement carries positive coefficients in all specifications. The estimated elasticity of exchange rate is about 7.6, indicating that a RMB appreciation by 10% decreases the likelihood of entering the export markets around 70%. Furthermore, the estimated coefficients of the interaction terms between exchange rate and firm productivity is negative and significant at 1% level, which suggests that the productivity growth of firms can decrease the negative effect of exchange rate appreciation on export market entry.
Results of the interactions reported in columns (4)-(6) of Table 7 suggest that the effect of exchange rate movements on export market entry is stronger for private firms. Younger firms and non-eastern firms, which means that the private-owned enterprises, younger firms and non-eastern firms are less motivated to enter into export markets than other firms when facing exchange rate appreciation. In a nutshell, we get the following findings. First, exchange rate movements have a negative impact on firms' export market dynamics. An exchange rate appreciation increases the likelihood of export market exit and decreases the probability of export market entry. This finding is in line with the results of relevant literature (Ilmakunnas and Nurmi, 2010;Alvarez and López, 2008;Goerg and Spaliara, 2013). What is more, in comparison with foreign studies, we find that Chinese firms are more sensitive to the exchange rate appreciation than other countries. The most likely reason is that Chinese exporters almost have low mark-up because they are locked in the low-end of global value chain (Li et al., 2015).
Moreover, the effect of exchange rate appreciation on export market dynamics negatively relates to firm productivity, which indicates that higher productivity can help firms to overcome the adverse effect brought by appreciation. The possible explanation to this result is that higher productivity firms are more capable of affording the entry costs and absorbing the exchange rate appreciation (Roberts and Tybout, 1997;Bernard and Jensen, 2004;Berman et al., 2012).
Eventually, the influence of exchange rate appreciation on export market dynamics is stronger for private-owned firms, but weaker for older firms and eastern firms. The fact that private firms suffer more from the appreciation can be explained by following two possible reasons: first, the POEs are fully exposed to market competition, whereas the SOEs are protected by government (Zhang et al., 2003); second, the SOEs usually export low-elasticity products which aren't sensitive to exchange rate movements (Shi et al., 2008), by contrast, the POEs generally export low-end products which are vulnerable to exchange rate movements. The higher hazard of exiting and the lower propensity of entering for younger firms caused by exchange rate appreciation may be explained by their poor market experience and lower financial ability. The weaker response of eastern firms to exchange rate appreciation on export market dynamics can be attributed to the high agglomeration in eastern China (Yu, 2014).

Extensions and robustness checks
In this section, we explore a few scenarios of extensions and robustness checks to the relationship between exchange rate movements and export market dynamics.

Extensive margins
The relationship between exchange rate movements and export market dynamics may be affected by the adjustment of exporters in extensive margin as well as intensive margin, thus exporters may be more able to survive in export markets when exporting many products to many destinations and exporters who export single product or export to single destination may be more vulnerable in export markets (Hummels and Klenow, 2005). We identify the role of extensive margins by interacting the number of products and the number of destinations with exchange rate movements.
The results reported in Colum 1-2 of Table 8 show that the expanding of export variety and destinations increases firms' possibility of entering foreign markets and decreases risk of exiting.
Besides, the firms who export more products to more destinations are more likely to enter but less likely to exit from the foreign markets and such in the period of exchange rate appreciation because of their richer exporting experience and higher risk resistance capacity.

Import demand elasticity
The responses of exporters to exchange rate movements may be different across goods with different import demand elasticity. We discriminate all HS goods as high-elasticity goods and low-elasticity goods according to Broda et al. (2006) and explore the heterogeneous effects of exchange rate movements on export market dynamics for goods with different import demand elasticity by interacting the exchange rate movements with the dummy of high-elasticity goods which equals one if the import demand elasticity of that goods surpasses median. The results displayed in Colum 3-4 of Table 8 suggest that a RMB appreciation is associated with a lower probability of entering and a higher probability of exiting for firms that export goods with high elasticity.   Note: Robust t-statistics in parentheses. *** p<0.01, ** p<0.05, * p<0.1.

Different destinations
The exporters' reaction to exchange rate movements may relates to the destinations they export.
Since non-OECD countries are likely to conceive more exchange rate fluctuations and inflation volatility compared with OECD countries (Li et al., 2015). Column 5-6 of Table 8 report how the exporters' reaction to exchange rate appreciation differs across different markets by adding the interaction term between exchange rate movements and the dummy of non-OECD countries. The results indicate that firms are less likely to enter the non-OECD countries than OECD countries and firms who export to the non-OECD countries enjoy higher hazard of exiting during RMB appreciation.

The role of U.S. dollar peggers
The RMB exchange rate was pegged to the US dollar before 2005, which indicates that the inflation is the only source of exchange rate movements between RMB and U.S. dollar.
Meanwhile, U.S. and other U.S. dollar peggers (e.g. Hong Kong) as defined in Klein and Shambaugh's (2006) are the major destinations of Chinese exports. The sticky pricing of goods exported to the dollar peggers inevitably exerts an influence the effects of exchange rate movements on firms' export market dynamics. We evaluate its effects of by interacting exchange rate movements with the dummy of U.S., Hong KONG and another U.S. dollar peggers before 2005. We find in column 1-2 of Table 9 that firms are more likely to enter the markets of U.S. dollar peggers and firms who export to U.S. dollar peggers are less likely to exit in the period of RMB appreciation.

The liberalization of foreign trading rights
The foreign trading rights were restricted to Chinese state-trading enterprises before July 2004.
Trading rights have been fully liberalized by the enforcement of The Revised Foreign Trade Law in July 2004 which provides for trading rights to be granted automatically through a registration process for all domestic and foreign enterprises and individuals. The liberalization of foreign trading rights would undoubtedly stimulate a great number of firms to enter export markets and therefore lead to more intense export market competition as described in Table 2.
To capture the impact of and foreign trading rights liberalization, we include the interaction term between exchange rate movements and the year dummy, which takes value one over the period 2005-06, and zero otherwise. The results reported in column 3-4 of Table 9 show that firms are more likely to start their exports after the liberalization of foreign trading rights even in the period of exchange rate appreciation and firms are more likely to exit under such circumstance because of more intense competition. Note: Robust t-statistics in parentheses. *** p<0.01, ** p<0.05, * p<0.1.

Alternative definition of export market dynamics
Our empirical results may be sensitive to the identification of export market exit, entry and survival. We therefore redefine export market exit, entry and survival by following the approach adopted by Goerg and Spaliara (2013) in column 1-2 of Table 10. Exiter is redefined as the firm exported in t-1 and t-2 but not in t. Entry is redefined as firm exported in t-1 and t-2 but not in t.
Furthermore, owing to the higher possibility of exiting the export market during the first year as shown in section 2, we have also tried excluding the one-year observations (with duration =1) from the sample in column 3-4 of Table 10. As can seen from column 1-4, the results basically remain unchanged compared to the baseline analysis. Note: Robust t-statistics in parentheses. *** p<0.01, ** p<0.05, * p<0.1.

Alternative measures of productivity
Since productivity is an important determinant of export market survival as discovered by (Girma et al., 2004;Ilmakunnas and Nurmi, 2010;Askenazy et al., 2011), we check the robustness of our results by employing productivity estimated by solow residual in column 5-6 of Table 10 and measured by labor productivity in column 7-8 of Table 10 which is defined as added-value against the number of employees separately. We observe that the results are still in line with our previous findings in section 4.

Alternative estimation approaches
The empirical results may suffer from the impropriate estimation method. In order to avoid potential biases related to this, we adopt several alternative estimation approaches including: (1) the Probit model, which estimates the odds ratio rather than hazard ratio, (2) the truncated regression model to address the concern of left truncation as we mentioned in section 2.

Concluding Remarks
We in this paper shed light on the relationship between foreign exchange rate fluctuations and firms' export market dynamics using a firm-level production data from China National Bureau of Statistics and a firm-level trade data from China Customs over the period of 2000-2006. We adopt a discrete-time survival model in our empirical investigation and further execute several extensions and robustness checks to the baseline results. The main results of the paper can be summarized as follows: First, an exchange rate appreciation increases the likelihood of export market exit, reduce the capability of export market survival and decreases the probability of export market entry. Second, high productivity firms are less likely to exit from export markets and more likely to enter and survive in export markets in the period of exchange rate appreciation. Third, exchange rate appreciation decreases the likelihood of export market entering and increase the likelihood of export market exiting more for private-owned firms, young firms, ordinary trade firms and non-eastern firms. Finally, other sources of heterogeneity, such as extensive margins, import demand elasticity, different destinations, U.S. dollar peg, and the liberalization of trading rights also matter, but our results are robust to alternative redefinition of dynamics, alternative productivity measures and alternative estimation approaches.
Our study gains some policy implications. First, exchange rate appreciation reduces the chance of entering export markets and increase the risk of failure in export markets, thus the government should do their best to avoid large exchange rate movements. Second, the entry and exit of high productivity firms are less affected by the exchange rate appreciation. thus exporters should increase their trade competitiveness by improving productivity through technology innovations, which should be encouraged and protected by government through IPR protection, tax mitigation, subsidy. Finally, the effects of exchange rate movement on export market dynamics relate to other firm heterogeneity. The policies implemented by government should take these heterogeneous effects into account in order to improve their adaptability.