Exchange Rate and Country’s Export Competitiveness: An Empirical Discourse Analysis

The objective of this study is to empirically analyze the discourse on correlation between exchange rate and country’s export competitiveness, which is one of the dominant discourses having been continuously and extensively reproduced in the Thai society by the authorities from academia, the public and private sectors. The study analyzes the time-series data of the exchange rate, overall exports, exports of agricultural products, and exports of industrial products by employing advanced statistical analysis, regression, and the Johansen Cointegration Test. Through regression, we find that exchange rate is negatively related to overall exports, exports of agricultural products and exports of industrial products. On the contrary, Johansen Cointegration Test does not demonstrate any long-term relationship between exchange rate and the three variables of export. Thus, the claim that the appreciation of domestic currency will negatively affect the country’s export competitiveness, whether in overall, agricultural products or industrial products, is not a defect in good faith but another example of the dominance of oriented discourse, which gives way for the elite to take advantage as a camouflage in the form of knowledge and truth, while others in the society never suspect nor argue about it.


Introduction
1.1 On "Discourse" and "Discourse Analysis" "Watakarma" is a terminology in social sciences and humanities, which Professor Somkiat Wanthana of the Social Sciences Faculty of Kasetsart University translated into Thai from the "Discourse" in concepts which scattered among the works of Michel Foucault, an important poststructuralist French philosopher more than 20 years ago (Wongsurawat, 2010, p. 3).
In the same way that Foucault defined "Discourse" different from people's perception in everyday life, that is verbal communication, speech or conversation (Heywood, 2000, p. 87), Wantana formulates "discourse" similar to the original term in English, as well as offers the meaning beyond simply combining the words "speech" (in Thai, "wata") and "action" (in Thai, "karma") (Jermsittiparsert, 2010, p. 187).
More than a decade later, the understandings on "social text" and "social reality" constructed and maintained through languages (Alvesson & Karreman, 2000, p. 1126), have finally been seriously processed and extensively published among Thai academics.Such process has been conducted through the writings and studies by Professor Chairat Charoensinolarn (2006, p. 19-20) of the Faculty of Political Science, Thammasart University, as the system and process of creating/producing identities and meanings for various objects in the society which encapsulates us, covering knowledge, truth, power, or the identity of our own, as well as things we have created, maintained, and are accepted by society at large, and finally turned into "dominant discourse".
The process of turning something into dominant discourse involved the use of "power" over refining/cleaning in the form of "specialized knowledge/expertise" through the traditional rules and practices of the "experts/authorities" into the "academic discourse" on the specific matter.This seems like a process that is natural, neutral, universal, objective and harmless to any person (Charoensinolarn, 2008, p. 421-422).It is similar to labeling the complex strategical situation in a particular society (Charoensinolarn, 2006, p. 25-26) in order to impose/turning, something which is not true, into the stage of "knowledge" and "truth" eventually.
As a critical analysis method, discourse has become a very popular topic and expanded to cover various academic fields within a short time after Foucault (Vighi & Feldner, 2007, p. 141).It appears in both textbooks and science-oriented discussion forums in different and various meanings (Jorgensen & Phillips, 2002, p. 1).However, the process is often without establishing a definition, but only an assumption, resembling the "discourse analysis" (Cheek, 2004(Cheek, , p. 1141)), which in gist, is the study and research on the process, steps, sequence of events, and different details, in order to establish an identity and meaning to things that encapsulate us in the society in the form of a discourse, including the practices and impacts of a discourse on the specific matter in the nature of relationships (Charoensinolarn, 2006, p. 27-28).
From the development of linguistics in early age to anthropology in the 1960s and psychology from in the early 1970s (Dijk, 2007), or even the present interdisciplinary (Schiffrin, Tannen & Hamilton, 2001, p. 1), the discourse analysis of more than 40 approaches has been associated with the study and analysis of the language (Sarangi, 2009;cited in Crichton, 2010, p. 13) with qualitative methodology (Hodge, Kuper & Reeves, 2008).Such methodology extensively brought about questions on the negligence on scientific reasonableness (Freshwater, Cahill, Walsh & Muncey, 2010, p. 502) in describing the linguistic structure, the duty of the texts, and revealing the relationship between the text and social context (Stillar, 1998, p. 14).Such methodology has been more emphasized than proving the truth of the context (Charoensinolarn, 2006, p. 29) with empirical data, statistics or quantitative methodology, which is driven to become only a way of (flawed) thinking of scientific social sciences or the school of logical positivism (Charoensinolarn, 2008, p. 432).

Discourse on the Relationship between "Exchange Rate" and "The Country'S Export Competitiveness"
The current literature review cannot find a conclusion associating with person, time, place, as well as the context of the beginning of the discourse on the relationship between "exchange rate" and "the country's export competitiveness", whether as "overall export" or classified into "export of agricultural products" and "export of industrial products".
It has been widely known that this matter closely relates to an economic theory which indicates that competitiveness of domestic and foreign operators closely relates to price, which in turn is determined by the exchange rate (Schembri, 1989;Piana, 2001).With this principle, people believe that whenever the currency weakens, export will increase and in the mean time, import will decrease (Chaiboonsri, 1999;Sakunbongkot, 2001;Kaewmanee, 2007;Patchimnan, 2011, p. 217).
The abovementioned assessment has been held upon and reproduced in academia, government and private sectors in Thailand, including Kriengsak Charoenwongsak (n. d.)This has led to the mobilization of financial policies by the government so as to provide assistance, which has brought about questions, similar to those raised by Ammar Siamwalla (Prachachat, 2010, October 21), Honorary Academician of Thailand Development Research Institute, who criticized whether such proposition is appropriate, who receives the benefits, and eventually for what purpose it is done.

Objective of the Research
This research has been conducted in order to analyze the relationship between the exchange rate and the country's export competitiveness, covering both overall exports and the classification of exports of agricultural product and exports of industrial products.This will reveal that the discourse repetitively reproduced by the authorities is the logical and reliable truth or it is merely an erratic viewpoint, inaccurate, doubtful, and only full of efforts to impose the proposition an arrogant status, that is, to become knowledge as well as the truth.

Data and Variables
The researchers use time series data, which has been continuously and systematically collected, publicly available, reliable, and generally referred by authorities and agencies, covering the period of 172 months from August 1997 to November 2011.The variables covered are as follows: 1) Exchange Rate (ER2) is a reference exchange rate between the baht against the US dollar, according to the average exchange rates of commercial banks in Bangkok from the data of the Bank of Thailand (2001Thailand ( , 2012)).ER2 is a set of independent variables of overall exports, exports of agricultural products, and exports of industrial products; 2) Overall Exports (EXPORT2) are the statistical exports according to the balance of payment statistics released by the Bank of Thailand ( 2011), herein defined as the dependent variable of the exchange rate; 3) Exports of Agricultural Products (AGRI2) are the date from the Bank of Thailand ( 2011), herein defined as the dependent variable of the exchange rate; and 4) Exports of Industrial Products (INDUST2) cover four industry groups, including labor-oriented industries, high-technology industries, local-raw-materials-oriented industries, and other industries, whose data are collected by the Bank of Thailand ( 2011), herein defined as the dependent variable of the exchange rate.

Data Analysis
The mainstream of discourse analysis has specifically focused on the consideration of structure, strategy, and process (Dijk, n. d.) and the (re)production of discourses, covering every social activities (Gadavanij, 2006, p. 31).Foucault (1994, p. xiv) accepted that it was complicated and cumbersome, not easily to explain with clarity.Also, it has been with the invention of technical terms of specific meanings, far from the definition which is understandable, appears in the dictionary, or publicly used on a daily basis.This makes it more a process of searching for an expertise in encryption than, in academic language in old paradigm, a fight that language is a form of power, and the origin of knowledge and the truth (Charoensinolarn, 2008, p. 422).
The purpose of analysis moved from concreteness, of what is true or false (Heywood, 2000, p. 88), to abstraction, that is, the rules govern the texts and degrade the value of human beings to a "skeleton" only following/reinforcing/reproducing such rules (Charoensinolarn, 2006, p. 29).However, eventually, virtually all of the research, which is based on the assumptions of denying the existence of an objective reality, is left with only the fact that was constructed (Moreland, 2005, p. 79), that is, an amusing game with emptiness, caught in the turmoil of subjectivity.It is a prediction of rules that even self could not conclude that they are correct.
Furthermore, there was a drive away from/protest against "verification", by assuming that it is just the narrow-minded essence of the school of logical positivism (Charoensinolarn, 2008, p. 11-12), instead of borrowing such idea to support the explanation to make it more efficient, and to convince the masses to be enlightened and aware of the inconsistency of the logic/crevice of discourse, as well as the lack of natural course (Charoensinolarn, 2006, p. 8).
The Empirical Discourse Analysis or EDA has thus been developed in order to return to Foucault (1980, p. 53-54), similar to what he did with Friedrich Nietzsche, the existentialist German philosopher.It is the transformation of his new examination by borrowing the principle of empiricism so as to "falsify" the core of the texts, which are made up of variables, connecting with the "statistic" language and continuously and extensively reproduced without any doubt from members of the society from the beginning, using the data set, logics, and methodology similar to that established.This research employs the advanced quantitative methodologies, regression analysis so as to test the relationships between variables.Later, the Johansen Cointegration Test (1988, 1991) is employed to double-check such relationships whether they have cointegrations, using built-in computer software, instead of simply discussing it as a discourse or discursivity (Charoensinolarn, 2006, p. 3).

Exchange Rate
The time series data of the exchange rate shows that the rate started in August 1997 at 32.48 baht per U.S. dollar and ended in November 2011 at 30.96 baht per U.S. dollar.The exchange rate has swung continuously on a monthly basis.When considered as a whole, the exchange rate shows a tendency to decrease, with the highest of 53.81 baht per U.S. dollar in January 1998 and the lowest at 29.88 baht per U.S. dollar in August 2011.The regression analysis indicated in table 1 shows that time has the relationship in the same direction as the exchange rate with statistical significance of .01.In every month, the exchange rate of Thai baht is moving in the opposite direction for 0.07 baht per U.S. dollar.That means, the time variable can explain the exchange rate for 52.61 percent.Table 2 shows that time has in the same-direction relationship with overall exports with statistical significance at 0.01.In every month, the overall export moves in the same direction for US$86.81 million.The time variable can explain overall exports to 86.85 percent.The regression analysis shown in table 3 indicates that time correlates with exports of agricultural products with a statistical significance at the .01level.In every month, the value of agricultural exports changed in the same direction as time for US$8.46 million, that is, the time variable explains the exports of agricultural products for 73.62 percent.

Correlation between the Exchange Rate and Overall Exports
In the beginning, the regression analysis shows that the exchange rate and overall exports have the relationship at a statistical significance of .01 in the opposite direction.Any change of the exchange rate at any US$ 1 will result overall exports moving in the opposite direction for approximately US$ 844.83 million.The exchange rate is a variable that can explain the overall exports for 67.26 percent.

Correlation between Exchange Rate and Exports of Agricultural Products
The regression analysis in table 6 implies that the exchange rate and exports of agricultural products have the relationship at a statistical significance at the .01level in the opposite direction.For the change of the exchange rate of US$ 1, exports of agricultural products will change in the opposite direction for approximately US$ 87.74 million.The exchange rate is a variable that can explain the exports of agricultural products for 64.72 percent.

Correlation between Exchange Rate and Exports of Industrial Products
The regression analysis in table 7 denotes that the exchange rate and exports of industrial products have the relationship at a statistical significance of the .01level in the opposite direction.For the change of the exchange rate of every US$ 1, exports of manufactured goods changed in the opposite direction for approximately US$ 757.34 million.The exchange rate is a variable that can explain the exports of industrial products for 67.19 percent.At this time, the discourse on the relationship between the exchange rate and the country's export competitiveness, whether overall exports or the classification of exports of agricultural products and of industrial products, which have been created by the authorities that the strong baht would affect the price competitiveness, causing the exports in all aspects above to decline, is not always true.The regression analysis shows on the other hand that although the exchange rate actually has the relationship with all the three variables but in the opposite direction, contrary to the mainstream discourse abovementioned.
However, the relationship found at this point, with intentional use of the non-stationary series data in employing the regression analysis, may be only falsified relationships (Box & Jenkins, 1976), which requires a double check by the Johansen Cointegration Test.

Cointegration Test between Exchange Rate and Overall Exports
The Johansen Cointegration Test points that the exchange rate and overall exports have no long-term equilibrium relationship at the .05level of significance, that is, the relationship found in the regression analysis, is not only the opposite-direction relationship but simply a false relationship.

Conclusion
In summary, the advanced quantitative methodology empirically shows that there is no relationship between the minimum wage rate and the country's economic competitiveness both in foreign direct investment in industrial sectors and export (Jermsittiparsert, Sriyakul & Pamornmart, 2012b) and domestic private investment (Jermsittiparsert, Sriyakul & Pamornmart, 2012a).
The methodology also empirically shows that the exchange rate has no relationship with the export competitiveness, both in overall exports, exports of agricultural products, and exports of industrial products.
Using the same data set, the same logic, and the same methodology as those applied in the exchange rate created and connected to the country's export competitiveness, both overall exports and exports of agricultural products and exports of industrial products and claim regarding the price advantages comparing to the competitors in the direct correlation, show the academicians' superficiality and lack of empirical date proof, the authorities' lack of examination, as well as the private sector's lie for its own benefits.
Under the basis of the school of thought of structuralism which states that "each person is speaking with a purpose" (Chaika, 2008, p. 150), this may not be deemed a failure in good faith, but also depicts the efforts of establishing and growing a belief through the legitimacy of the authorities (Weber, 1997, p. 325) through the use of language.Discourse, including a reproduction of it, is such a powerful and efficient tool (Leeuwen, 2008, p. 105) to dominate and economically take advantage of and discriminate against people of different social classes, all of which are the root of Thailand's important structural problems.Such fights have been utilized by authorities to transform their own viewpoints into knowledge as well as truth, while other members do not have doubts or argue.

Recommendation
Although the answer found in the "Empirical Analysis of Discourse" is not what the discourse analyzers in the current paradigm is interested in, but the application of this new type of examination can fill the holes in the critic and empirically reveal the inconsistency of logic as well as the split of the discourse by "falsifying" the first hypotheses.Such a way is the clear beginning of the critical investigation of the texts that have been (repeatedly) produced by the authorities, until they become the familiar, often overlooked in society, and socially accept without questions or examinations.It is completely different from the process that is vague, challenging people to raise questions if such concepts are based on people's own interpretation.
The application of such methodology as a tool for monitoring and preventing the (re)production of the discourse, will not only create hopes that the public can access the clarification further and more easily, but also provokes a wave of criticism, which is the condition for developing the methodology later.
Figure 1.Exchange rate Figure 2. Overall exports

Figure 3 .
Figure 3. Exports of agricultural products

Table 1 .
Regression analysis of exchange rate and time Overall exports in August 1997 valued at US$ 4,954.96 million while in November 2011 totaled at US$ 15,496.37 million.The monthly change of the value of overall exports switched between increasing and deteriorating.When considered as a whole, the overall exports tend to increase, with the lowest at US$ 3,979.00 million in January 1999, and the highest at US$ 21,258.72 million in September 2011.

Table 2 .
Regression analysis of overall exports and timeThe time series data for exports of agricultural products shows that exports of agricultural products in August 1997 valued at US$ 525.81 million ended in November 2011 at US$ 1,818.38 million.The values of exports of agricultural products swung on a monthly basis.However, when considered as a whole, the exports of agricultural products have an increasing trend, from the lowest of US$ 320.14 million in April 2001 to the highest of US$ 2,498.48 million in March 2011.

Table 3 .
Regression analysis of exports of agricultural products and timeFigure 4. Exports of industrial productsThe regression analysis in table 4 reveals that time correlates the exports of industrial products at a statistical significance at the .01level.Every month, the value of manufactured exports changed in the same direction as time for US$ 78.16 million, that means, the time variable explains the exports of industrial exports for 87.53 percent.
goods has swung monthly.However, when considered as a whole, the values tend to increase, from the lowest at US$ 3,377.24 million in January B.E. 1999 to the highest at US$ 18,936.23 million in July 2011.

Table 4 .
Regression analysis of exports of industrial products and time

Table 5 .
Regression analysis of exchange rate and overall exports

Table 6 .
Regression analysis of exchange rate and exports of agricultural exports

Table 7 .
Regression analysis of exchange rate and exports of industrial products

Table 8 .
Johansen cointegration test on exchange rate and overall exports Cointegration Test between Exchange Rate and Exports of Agricultural ProductsThe Johansen Cointegration Test results in table9shows that the exchange rate and exports of agricultural products have no long-term equilibrium relationship at the .05level of significance.The relationship found in the regression analysis, is not only the opposite-direction relationship but simply a false relationship.

Table 9 .
Johansen cointegration test on exchange rate and exports of agricultural products Cointegration Test between Exchange Rate and Exports of Industrial ProductsThe Johansen Cointegration Test results in table10shows that the exchange rate and exports of industrial products have no long-term equilibrium relationship at the .05level of significance.The relationship found in the regression analysis, is not only the opposite-direction relationship but simply a false relationship.

Table 10 .
Johansen cointegration test on exchange rate and exports of industrial products