MEDIT N° 4/2001

COMPARATIVE ADVANTAGE AND UNCERTAINTY IN THE INTERNATIONAL TRADE OF MEDITERRANEAN AGRICULTURAL PRODUCTS:

INTRODUCTION

International competitive­ness is more often than not defined in terms of trade deficit, national productivi­ty growth, and the sectoral decomposition of output and their determinants. However, there is consider­able confusion on what is meant by the term competi­tiveness, particularly when it refers to a country. Haque (1995) defined international competitiveness as the abil­ity of a country to produce goods and services that meet the demand of inter­national markets, and si­multaneously maintain and expand the real incomes of its citizens. This definition suggests that when the unit of measurement is the county, indices of competi­tiveness based on trade on­ly depict the half picture of a country's competitive­ness. At the same time, the definition indirectly relates competitiveness to a gov-

AN EMPIRICAL ANALYSIS

ILIAS P VLACHOS (*)

ABSTRACT

This study applied the indicators of uncertainty on the international trade of agricultural products. Proxies of uncertainty were developed using the long-term variability of two comparative advantage indica­tors: the Revealed Symmetric Comparative Advantage index (RSCA) and the Revealed Competitiveness Index (RC). Variability was calcu­lated as the coefficient of variation (CV) for both RSCA and RC indices for Mediterranean basin countries over the period 1982-1997. The cv of RSCA was defined as Revealed Export Uncertainty (RXU) and the CV of RC index was defined Revealed Trade Uncertainty (RTU). Re­sults indicate that, given certain limitations, both RXU and RTU can be used as proxies of uncertainty. Referring to Mediterranean states, RXU and RTU produced striking results of the export performance in con­trast to RSCA and RC indices which exhibited low interpretability. Un­certainty indices appear to better depict the comparative position of Mediterranean basin countries.

RESUME

Cette etude a applique les indicateurs d 'incertitude sur le commerce in­ternational des produits agricoles. Des proxies d'incertitude ont ele developpees en utilisant la variabilite a long terme de deux indicateurs d'avantage comparatif: 1Indice d'Avantage ComparatiJ Symetrique Revele (RSCA) et 11ndice de Competitivite Revele (RC). La variabilite a ete calculee en tant que coeffiCient de variation (CV) tant pour les in­dices RSCA que RC pour les pays du pourtour Mediterraneen dans la periode 1982-1997. Le CV du RSCA a ete defini comme etant IIncerti­tude Revelee de l'Exportation (RXU) et le CV de l'indice RC a ete defini I1ncertitude Revelee du Commerce (RTU). Les resultats indiquent que, compte tenu de certaines contraintes, tant le RXU que le RTU peuvent etre utilises comme proxies de I 'incertitude. Se rejerant aux Etats mediterraneens, le RXU et le RTU ont produit des resultats jrappants de la performance des exportations contrairement aux indices RSCA et RC qui se sont averes d'interpretation plus difficile. Les indices d'incel'­titude semblent mieux decrire la position comparative des pays du pourtour Mediterraneen

or supernational associa­tions exposed, and remain­ing exposed, to secure inter­national competition, to se­cure a relatively high return on the factors of production and relatively high employ­ment levels on a sustainable basis'. The principle of compara­tive advantage provides an explanation of specialisa­tion and gains from trade. Comparative advantage also provides a base to yield predictions about the direc­tion and the terms of trade (Goldin, 1990). Ricardo was the first to advocate the no­tion of comparative advan­tage as a determinant of in­ternational trade in an at­tempt to explain why Portu­gal exported wine and Britain cloth. However, ac­cording to Ricardo's theory, comparative advantage is confined to labour produc­tivity thus, only one factor of production is considered. According to the Hecksch­er-Ohlin principle, coun­

ernment's policy in creating the conditions to make trade more competitive, e.g. by giving export incen­tives, or signing international trade agreements. It is this governmental intervention which makes the measure­ment of competitiveness more difficult to achieve. Furthermore, this level of measurement might not re­flect the competitiveness in industrial sectors such as food, where changes in the balance of trade and/or per­formance of the firms in the sector might determine competitiveness (Traill and Silva, 1996). For example, European Commission (994) defmes competitiveness as:

tries that are rich in certain factors will export goods that make use of those abundant factors intensively. In other words, the Heckscher-Ohlin principle states that comparative advantage arises from the different relative factor endowments of a country's trading. Haberler (1976, p. 4) remarked that "no sophisticated theory is necessary to explain why Kuwait exports oil, Bolivia tin, Brazil coffee and Portugal wine", suggesting that "natural resources" determine trade patterns. Thus, natural resource trade reflects a country's comparative advantage. Based on the above theories, it can be assumed that trade is a proxy for comparative advantage. The as­sumption is based on the argument that if Cl country is specializing in and exporting something in a competi­tive market economy, this must be its comparative ad-

"the capaCity of businesses, industries regions, nations

CO) Research Officer - Agricultural University of Athens, Greece -iliasvlachos@yahoo.co.uk

42

MEDlT N° 4/2001

vantage sector (Kilkenny and Nalbarte, 2000). It can be further assumed that a country with a comparative ad­vantage sector would struggle to create everything but uncertainty and vulnerability in the sector, conditions that might result to declining exports.

MEASURES OF COMPARATIVE ADVANTAGE

Amongst the measures of comparative advantage, net exports (exports plus imports), export growth, and the export market share can be easily computed. The Ex­port Market Share (XMS) index is calculated as:

(1)

where: XMS. is Export Market Share of country j; Xi is the value of! country's y exports of product i; Xiw is the value of total world exports of product i. However, XMS and its homologous indices are biased to the extent to which exports may fluctuate due to an unexpected increase (decrease) in prices, or due to a currency devaluation which will make products very at­tractive in the foreign markets. In such cases, XMS might not measure the comparative advantage of a spe­cific product but the short-term attractiveness of a coun­try's exports. Balassa (1965) was the first to develop an index in or­der to "reveal" comparative advantage using export per­formance ratios. The index, which will be called Ex­port-Revealed Comparative Advantage (XRCA) , is a ra­tio with numerator a country's share of exports of a par­ticular product to its total exports and denominator the commodity's share to world exports. Hence, XRCA is calculated by the formula:

Xij

XTj

XRCA,,= X"w IJ __

X 1W

(2)

where Xi' is the exports of country j of product i; ~, is the sum bf exports of country j; XiW is the world expbrts of product i; ~ is the World Total Exports. The XRCA ranges between zero and unity in case a country is not specialised in exports and from one to in­finity if it is specialised. Therefore, the index is static and asymmetric. The export shares that comprise the ReA ratio are in­fluenced by trade interventions such as protectionist barriers in export markets and anti-export bias in do­mestic trade policy that impose a certain bias on the re­liability of the index. Particularly, the bias can be con­siderable when referring to agricultural trade because of the widespread application of such policy measures in international trade. Therefore, the XRCA index is more likely to depict reality better where computed for

43

processed goods or manufactures where trade interven­tions to free liberation occur less frequently. There are a number of variations to the above RCA in­dex. For example, variations of this index can be devel­oped by altering the denominator of the ratio substitut­ing world by a specific group of countries (e.g. the in­dustrial countries, OECD, Mediterranean countries, etc.). Balassa (1965) again suggested RCA indices that took into account changes in the export performance ratio over time and in order to arrive at a dynamic mea­sure of comparative advantage. Laursen (1998) calculat­ed the so-called symmetric RCA (RSCA) in order to overcome difficulties of using RCA values in statistical models. RSCA is given by the formula:

l-XRCA .. RSCAij = 1 + XRCAIJ ,

IJ

(3)

Laursen (1998) found that RSCA is, on balance, better than other indices. RSCA take values from - 1 (when XRCA tends to infinite, which indicates absolute export advantage) to + 1 (when XRCA is zero and exports are minimal). Another approach to measure comparative advantage is the so-called "export-import" RCA (EIRCA), which is de­fined as a country's ratio of exports to imports of a par­ticular commodity to the ratio of world exports to world imports of that commodity. In EIRCA index, a value greater than one indicates a comparative advantage. Vollrath and Vo (1988) developed the "revealed com­petitiveness" (RC) index which incorporates both im­ports and exports. The RC is given by the formula:

or M;;

MTj

RCij = XRCA - MiW

Mrw

(4)

where Xi' is the exports of country j of product i; ~, is the sum bf exports of country j; XiW is the world expbrts of product i; ~ is the World Total Exports; Mi' is the imports of country j of product i; ~, is the sum! of im­ports of country j; Miw is the world iinports of product i; ~ is the World Total Imports.

DISCUSSION OF COMPARATIVE ADVANTAGE INDICES

It is evident from the above discussion that the empiri-

MEDIT N° 412001

cal measures of comparative advantage share some common characteristics: • First, the variable of comparative advantage should be defined and measured in relative terms. This is con­sistent to Ricardian and Heckscher-Ohlin models of in­ternational trade. A country's comparative advantage has to be measured in relation to the performance of another group of countries, being either the world or a super national association to which the county shares some common characteristics (e.g. the European Union or OECD countries, etc.). For example, XRCA relates to a country's exports to the world exports. • Second, measures of comparative advantage are heavily dependant on the availability of data, which usually refer to export and import values and quantities. Based on the available data, a sufficient number of studies have laid the foundations for the statistical mea­sures of comparative advantage (Le. Haley, 1987; Perkins, 1987; Vollrath, 1987; and, Vollrath and Vo, 1988). • Third, although comparative advantage is considered a dynamic concept, common measures of comparative advantage can only capture static instances of its evolu­tion. • Fourth, indicators are ex-post measures of compara­tive advantage. They are based on empirical data of past performance to reveal current and future perfor­mances.

REvEALED UNCERTAINTY

There is a pronounced overt relationship between comparative advantage and trade uncertainty. Stilwell (2000) acknowledged this adverse relationship between comparative advantage and uncertainty in a straightfor­ward manner by entitling his study as: "Dollar Down: Uncertainty Up". If a government' aim is to promote in­ternational trade, one way of achieving is by sustain the country's comparative advantages, thus reducing uncer­tainty about them. For example, Ibarra (1992) found that international trading agreements have reduced the severity of uncertainty. There is consensus that uncertainty has a negative im­pact on a country's critical macroeconomic indicators. Palkhiwala (1993) found that export instability has proven to be detrimental to the economic development efforts of developing countries. Abbasi (1991) found that uncertainty associated with fluctuations in oil ex­port earnings affect adversely the economic growth in the oil-based economies of the Middle East, through the negative effect on investment, government expendi­tures and domestic output. Smets (1993) and Wei (1997) found that uncertainty substantially delays and reduces foreign direct investment. Gallagher (1998) found that trade uncertainty adversely affects marketing firms and commodity trade, thus its reduction would result in im­proving world welfare. Ghirmay (1997) examined em-

44

pirically the long-term and short-term relationships be­tween exports and economic growth in a sample of 30 developed and developing countries and found some indications that export instability may have negative ef­fects in developing countries by creating uncertainty in their capacity to import capital goods and intermediate inputs. Marrewijk and Bergeijk (1993) reported an even more dramatic consequence of uncertainty in interna­tional trade, an adverse specialisation of a small econo­my in the production of the goods in which it has a comparative disadvantage. However, there is no consensus in the literature on how to measure uncertainty. Standard deviations and coeffi­cients of variation are often used in order to measure volatility which appears to unveil uncertainty. For ex­ample, Edison and Melvin (1990) reviewed the litera­ture on the impact of volatility of exchange rate on trade. The majority of proxies of volatility were the standard deviations of corresponding variables. The General AutoRegressive Conditional Heteroskedastic (GARCH) model, which is based on the assumption that the variance of the error terms is not constant over time, is an alternative to construct uncertainty proxies. For the purpose of this study, it was assumed that the volatility in a country's comparative advantage reveals its trade uncertainty. Therefore, coefficients of variation of the RCA indices were used as proxies of trade un­certainty. Two such proxies were calculated: the Re­vealed Trade Uncertainty (RTU) and the Revealed Ex­port Uncertainty (RXU) based on the values of the vari­ables RC and XRCA respectively. The CV was preferred than standard deviation for its power to compare the variability of two or more distributions when figures are of quite different orders of magnitude. For example, Yanagida and Tian (1996) used coefficient of variation to pinpoint countries with relatively stable RCA and RC indices. The RTU was calculated using the absolute val­ues of RC in order to arrive at comparative figures. The coefficient of variation (CV) is given by the formu­la:

CV= S tan dard Deviation

Mean

n-l ----- *100 (5)

io::::n

Ix; i-I

Furthermore, as countries exported a variable number of products, the coefficients of variation were adjusted to this variable. Both RSCA and RC indices were calculated using avail­able data from FAO database. RSCA index was calculat­ed using formula (3) and RC index using the formula (6), respectively. Products with export values less than US $ 1,000 for the last two years (1996 and 1997) were

"'IEDIT N° 4/ 200 I

excluded from the analysis in order to facilitate compu­tation tasks. The ratio CV over the number of products was consid­ered a better estimation of a country's volatility, since the volatility in the comparative advantage of a single product or a product ca tegory wo uld not signify a country's overall trade uncertainty. Therefore, the Re­vea led Export Uncertainty index (RXU) was calculated as:

11

)' C~?SCA .. RX~ = i=1 IJ

n (6)

where RXU. is the Revealed Export Uncertainty index of country j; CV the coefficient of variation of the index RSCA

ii; n the number of products exported from coun­

try j. RXU, like CV, takes values from zero (0), which in­dicates an absolutely certain, stable trade environment to 100% or more, which indicates high levels of exports uncertainty . RXU was calculated as the volatility of a country 's com­pa rative advantage, expressed by the RSCA index. Volatility in a country's comparative advantage can be considered to indicate uncertainty in exports. For ex­ample, significant flu ctuations of the RSCA index indi­ca te instability of exports, which , to a large extent , might be attributed to an ineffective export policy. As a result, exports instability will be reflected with high val­ues of RXU. The Revealed Trade Uncertainty (RTU) was calculated as :

11

R~ =

)' CVRC i=1 "

n (7)

where RTUi is the Revealed Trade Uncertainty index of country j; CV the coefficient of variation of the index RCii ; n the number of products traded (imported/ ex­ported) from country j. RT , like RXU, takes values from zero (0) , which indicates no uncertainty in trade, to 100% or more, which indicates complete uncertainty in trade. RTU is calculated as the volatility of a country 's trade competitiveness, which indicates trade uncertainty. RTU is based on the variability of RC index which in­corporates imports as well exports in order to arrive at a better estimation of a country's competitiveness . Va lues of RT and RXU were calculated for every coun­try of the Mediterranean basin for the period 1982-1997. Having calculated RXU and RTU, it was able to rank countries according to their degree of trade uncertainty.

RESULTS

Table 2 presents the export value for the Total of Agri-

45

Table 1 A Sample of the 423 Products Included in the Analysis.

Abaca Cherries Milled Paddy Rice Almonds Chicory Roots Mushrooms Animal Oil Cigarettes Oats, Rolled Apple iuice Concentrated Cigars Cheroots Oil of Castor Beans Apples Cinnamon (Canella) Oil of Olive Apricots Cocoa Beans Oil of Olive Residues Areca Nuts Coconuts Olives Artichokes Coffee Extracts Olives, Preserved Asparagus Cotton Lint Onions Avocados Dry Whey Pet Food Bacon-Ham of Pigs Figs Pig Meat Bananas Flax Fibre and Tow Pineapples Bananas Flax Fibre Raw Pineapples, Canned Barley Flax Tow Waste Pistachios Barley Flour and Grits Flour of Fruit Plums Beans, Green Flour of Maize Pork Beef and Veal Garlic Potatoes Beer of Barley Ginger Sausages Beeswax Goat Meat Silk Berries Nes Goose Meat Sisal Beverages Grape Juice Sorghum Blueberries Grapes Sour Cherries Bovine Meat Hen Eggs Soya Sauce Bread Prep. Soybeans Breakfast Cereals Honey Spinach Butter Hops Strawberries Butter of Cow Milk Kiwi Fruit Sunflower Seed Cabbages Lemons and Limes Tobacco Canary Seed Lentils Tomatoes Canned Mushrooms Lettuce Vani lla Carobs Macaroni Vegetables Frozen Carrots Mangoes Watermelons Cauliflower Margarine Wine Cereals Milk Fresh Yoghurt

Source: FAO (www.fao.org)

cultural products fo r the Mediterranean countries. France is by far the most export-oriented Mediterranean country with about 40 billion US $ value in 1997 fol­lowed by Italy and Spain ($ 15 bn). Turkey is the only non-European Union (non-EU) state with export value comparative to EU countries and well above Greek and Portuguese export value. EU states and Turkey show a steady increase of their export value for the sum of agri­cultural products (Figure 1) . Syria, Morocco , Cyprus, and the rest Mediterranean countries follow with less

Total of Agricultural Products

~ 45$ § 40$ .. :2 35$ A / --..

/'-" '¥' Q.) 30$

~ :::J

~ 25$ > /' 1:: 20$ 0 ./ Cl.

~-x 15$ UJ ---10$ --------5$ ~~"- -0$

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97

11 -+- France - Italy Spain -><- Turkey --- Greece -+- Portugal

Figllre 1 - E.:\porl Valu e q/To/a l q( AgriclIltura l Prodlfclsji"Onl [:.:U Nledilerraneall COIfIl ­(res: Valu e ill million US dol/aI'S du rillg (be period 1982-7997.

MEDIT W ,,/ 200 1

Table 2 Exports of Total Agricultural Products from Mediterranean countries during the period 1982-1997: Value in million US dol/ars.

Country/ 1982 1983 1984 1985 1986 1987 1988 Year

France 15,833 16,035 16,011 16,521 19,871 23 ,650 27 ,227 Italy 5,637 5,127 5,330 6,069 6,787 7,906 8,775 Spain 3,150 2,987 3,476 3,413 4,284 5,867 6,723 Turkey 2,588 2,409 2,395 2,211 2,319 2,588 3,034 Greece 1,330 1,549 1,574 1,381 1,822 2,003 1,420 Portugal 408 463 509 462 524 616 726 Syria 291 279 412 206 237 173 179 Morocco 375 364 311 367 450 457 572 Cyprus 204 158 196 149 188 205 197 Tunisia 158 107 152 141 161 194 193 Egypt 673 726 756 662 669 673 514 Lebanon 169 148 125 105 120 120 141 Jordan 201 145 135 122 136 121 97 Libya - - - - 1 2 2 Algeria 72 38 48 57 27 30 30 Malta 23 23 25 25 26 33 31

Note: Countries are ranked by the export value in 1997.

significant exports (Table 2) . The XRCA, RSCA, and RC values for the toral of agricul­tural products of all Medite rranean countries are shown by Table 3 , Table 4 and Table 5 , respectively. As shown in the values of XRCA fo r Cyprus are well about the threshold of one (1) which indicates export special­isation. The same stands true for about ha lf of the coun­tries studied (Greece, Syria, Lebanon, Turkey, Morocco, Spa in, France, and Egypt, subseque ntl y). Italy and Por­tugal appea r to have an export comparative disadvan­tage, since their XRCA values were between 0.6 and 0.8 for the entire period 1982-1997 (Figure 2). RSCA values reveal the same picture as this index is based on XRCA values (formula 3). Based on the values of RC index, Cyprus presents the highest revea led competitiveness among Mediterranean countries (Table 5). Greece and Turkey fo llow Cyprus with their RC va lues being at l.58 and l. 42, in 1997, re­spectively. Spain and France have lower but positive RC

]989

28,439 9,225 6,523 2,848 2,452

764 459 520 221 208 532 141 103

1 35 28

et U a: x

1990 1991 1992 1993 1994 1995 1996 1997

33,432 32,776 36,282 33,331 34,947 40,722 40,402 38,502 11 ,135 11 ,998 13,046 11 ,905 13,285 14,587 16,889 15,735 7,826 8,900 9,469 9,756 10,990 13,190 14,964 15,119 3,120 3,748 3,420 3,633 4,034 4,301 4,700 5,206 2,474 2,768 3,220 2,628 2,944 3,341 3,657 3,039

925 1,038 1,141 929 1,095 1,325 1,487 1,464 740 637 625 682 797 752 853 1,037 647 670 581 509 598 780 896 832 300 328 352 335 419 621 832 740 282 477 338 347 523 468 322 530 427 391 401 360 553 536 521 442 123 144 133 115 114 116 124 144 111 195 176 185 172 227 183 139

63 49 28 36 25 49 46 48 50 54 76 97 35 108 137 45 36 40 45 38 20 22 31 39

4.00 -r--------------------, 3.50

3.00

2.50

2.00 ~ 1.50 ~--..;; ... i:=-:;:--.......... ~==-... ,~

1.00

0.50

I--~~~~-------------------+

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97

Year

I - Greece --+-- Spain - France Italy --Portugal I

F(~lIre 2 - X NCA 11lde:~,: q/ I:.:U ilfech/err(fJletl1l Slales/or To/a/ Ag ricultural Products: /982· /997.

Table 3 XRCA index of Total Agricultural Products: Mediterranean countries during the period 1982-1997.

COLlntry/ 1982 1983 198-1 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Year

Cyprus 3.20 2.79 2.98 2.93 3.50 3.27 2.76 2.83 3.38 3.65 3.75 4.27 4.77 5.84 6.85 7.22 Greece 2.69 3.02 2.83 2.83 3.02 3.04 2.58 3.30 3.28 3.39 3.54 3.44 3.45 3.49 3.73 3.30 Syria 1.25 1.27 1.94 1.18 1.67 1.27 1.33 1.56 1.88 1.98 2.12 2.40 2.47 2.19 2.44 3.22 Lebanon 2.02 1.85 177 1.79 1.91 1.88 2.31 2.96 2.55 2.81 2.31 1.86 170 1.62 1.39 2.45 Turkey 3.93 3.67 2.92 2.60 2.91 2.52 2.59 2.50 2.58 2.93 2.44 2.62 2.45 2.30 2.32 2.42 Morocco 1.58 1.52 1.25 1.59 1.74 1.62 1.57 1.59 1.64 1.66 1.53 1.52 1.66 1.91 2.16 2.21 Spain 1.35 1.31 1.28 1.32 1.51 1.71 1.65 1.50 1.51 1.57 1.53 1.80 1.66 1.62 1.68 1.84 France 1.49 1.53 1.49 1.58 1.56 1.64 1.59 1.57 1.61 1.52 1.52 1.66 1.54 1.56 1.53 1.55 Egypt 1.88 1.97 2.10 1.67 2.14 3.28 2.41 2.05 1.77 1.14 1.38 1.28 1.76 1.80 1.69 1.37 Tunisia 0.69 0.49 0.73 0.76 0.85 0.90 0.80 0.72 0.86 1.37 0.87 1.02 1.24 0.99 0.67 1.16 Jordan 2.34 2. 19 1.55 1.45 1.74 1.28 0.94 0.94 1.12 1.83 1.52 1.64 1.33 1.48 1.15 0.92 Italy 0.67 0.61 0.63 0.72 0.65 0.67 0.68 0.67 0.70 0.75 0.77 0.78 0.76 0.72 0.77 0.80 Portugal 0.84 0.86 0.85 0.76 0.68 0.67 0.66 0.61 0.61 0.68 0.66 0.67 0.67 0.67 0.69 0.77 Malta 0.50 0.57 0.55 0.58 0.48 0.55 0.43 0.33 0.34 0.34 0.31 0.31 0.14 0.16 0.22 0.32 Libya - - - - 000 000 0.00 0.00 0.05 0.05 0.03 0.05 0.04 0.06 0.05 0.06 Algeria 0.05 0.03 0.04 0.04 0.03 0.04 0.04 0.04 0.04 0.04 0.07 0.10 0.04 0.12 0.12 0.04

Note: Countries are ranked by the XRCA index in 1997.

46

,\lEDIT N° 4 200 1

Table 4 RSCA index of Total Agricultural Products: Mediterranean countries during the period 1982-1997.

Country/ 1982 1983 1984 1985 1986 1987 1988 Year

Cyprus -0.52 - OA7 - 0.50 - OA9 - 0.56 - 0.53 - OA7 Greece - OA6 - 0.50 - OA8 - OA8 - 0.50 - 0.51 - OA4 Syria -0.11 - 0.12 -0.32 - 0.08 -0.25 - 0.12 - 0.14 Lebanon - 0.34 - 0.30 -0.28 - 0.28 - 0.31 - 0.30 - OAO Turkey -0.59 -0.57 - OA9 - OA4 - OA9 - OA3 -OA4 Morocco - 0.22 -0.21 -0.11 -0.23 - 0.27 - 0.24 - 0.22 Spain -0.15 - 0.13 -0.12 - 0.14 -0.20 - 0.26 -0.25 France - 0.20 -0.21 -0.20 - 0.23 - 0.22 - 0.24 - 0.23 Egypt -0.31 - 0.33 - 0.35 - 0.25 - 0.36 -0.53 - OA1 Tunisia 0.18 0.34 0.15 0.13 0.08 0.05 0.11 Jordan - OAO - 0.37 - 0.22 - 0.18 -0.27 -0.12 0.03 Italy 0.20 0.24 0.23 0.16 0.21 0.19 0.19 Portugal 0.09 0.07 0.08 0.14 0.19 0.20 0.21 Malta 0.34 0.28 0.29 0.27 0.35 0.29 OAO Libya 1.00 1.00 0.99 Algeria 0.90 0.94 0.93 0.92 0.93 0.93 0.93

Note: Countries are ranked by RSCA index in 1997.

values which indicates that these countries are compet­itive on trading agricultural products. On the contrary, Italy and Portugal appear to be less competitive, as their RC va lues were below zero for the entire period from 1982 to 1997 (Figure 3) . From Figure 2 and Figure 3, it can be observed that XRCA and RC ind ices produced the same rank of EU states . This is confirmed by Figure 4 which plots EU states according to their indices of XR­CA and RC for the total of agricultural products in 1997. Greece presents the highest values of both XRCA and RC indices fo llowed by Spain and France, while Italy and Portugal lag behind particularly rega rd ing RC in­dex. Table 6 presents the ranking of Mediterranean coun­tries according to their RXU and RTU indices. Countries in Table 6 are ordered by their RTU rank Table 6 also presents the sum of the coefficients of variation for XR-

1989

- OA8 - 0.53 - 0.22 -0.50 - OA3 -0.23 - 0.20 -0.22 - 0.34

0.16 0.03 0.20 0.24 0.50 1.00 0.93

>< Q)

"C .= u II:

]990 1991 1992 1993 1994 1995 1996 1997

- 0.54 - 0.57 - 0.58 - 0.62 - 0.65 - 0.71 - 075 - 0.76 - 0.53 -0.54 -0.56 - 0.55 - 0.55 - 0.55 - 0.58 - 0.54 - 0.31 -0.33 - 0.36 - OA1 - OA2 - 0.37 - OA2 - 0.53 - OA4 - OA8 - OAO - 0.30 - 0.26 - 0.24 -0.16 - OA2 - OA4 - OA9 - OA2 - OA5 -OA2 - 0.39 - OAO - OA1 - 0.24 - 0.25 - 0.21 -0.21 -0.25 - 0.31 -0.37 -0.38 - 0.20 - 0.22 -0.21 -0.29 - 0.25 - 0.24 - 0.25 - 0.29 - 0.23 - 0.21 -0.21 - 0.25 - 0.21 - 0.22 - 0.21 -0.22 - 0.28 -0.06 - 0.1 6 - 0.12 - 0.27 - 0.29 - 0.26 -0.16

0.08 -0.16 0.07 - 0.01 - 0.11 0.01 0.20 -0.07 - 0.06 - 0.29 - 0.20 - 0.24 - 0.14 - 0.19 -0.07 0.04

0.18 0.14 0.13 0.13 0.13 0.16 0.13 0.1 1 0.25 0. 19 0.21 0.20 0.20 0.20 0.18 0.13 OA9 0.50 0.53 0.52 0.75 073 0.64 0.51 0.91 0.91 0.94 0.90 0.93 0.88 0.90 0.88 0.93 0.91 0.87 0.81 0.92 0.78 0.78 0.92

2.50

2.00

1.50

1.00 .

0.50 ·

- 0.50 "'_/----- ---------1.00 "------------------'

B2 B3 B4 B5 B6 B7 BB B9 90 91 92 93 94 95 96 97 Year

I - Greece -+-Spain --<- France Italy - Portugal I

Figllre:1 - lie IlIdex olEe ,l/ediferrCllle{1I1 St{ftpsJiJr Tot{fl A,~ ricllltllml Prodllcts: 7982-1997.

Table 5 RC index of Total Agricultural Products: Mediterranean countries during the period 1982-1997.

Country/ 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 ]997 Year

Cyprus 2.00 1A4 1.77 1.67 2.17 2.08 1.77 1.88 2.37 2.70 2.94 2.57 2.91 3.65 4.04 4.09

Greece 1.60 1.86 1.71 1.61 1.52 1.30 0.85 1.73 1.71 1.96 2.09 1.95 1.76 1.79 2.11 1.58 Turkey 3.64 3A2 2.35 2.03 2.31 1.74 1.98 1.51 1.54 2.15 1.68 1.79 1.63 1.17 1.28 1A2 Syria 0.07 -0.37 0.38 -0.50 0.07 -0.30 - 0.55 - 0.92 -1.38 -0.70 0. 11 0.38 0.56 0.34 0.64 073 Lebanon OA6 0.52 0.17 -0.23 - 0.07 - 0.36 - 0.22 - 0.19 - 0.09 0.61 - 0.30 - 0.22 - 0.34 - 0.32 - OAO 0.68 Spain 0.37 0.24 0.24 0.30 0.28 0.72 0.68 0.57 0.57 0.57 0.52 0.55 OAO 0.24 OA6 0.61 France 0.59 0.56 0.50 0.57 0.51 0.61 0.60 0.62 0.67 0.56 0.55 0.58 OA4 OA5 OA7 OA7 Morocco 0.10 -0.09 - 0.59 - 0.10 0.12 0.16 0.15 0.15 OA6 OA1 -0.05 - OA4 -0.07 - OA8 - 0.16 0.04 Tunisia -0.31 -0.88 -0.60 - 0.59 - 0.50 - 0.30 -091 - 0.84 -0.35 OA6 - 0.05 0.05 0.09 - 0.54 - 0.53 - 0.20 Italy -0.58 -0.69 -0.63 -0.69 - 0.83 - 0.77 -073 - 0.71 - 0.63 - 0.63 -0.54 - 0.68 -0.68 - 0.56 -0.62 - 0.57 Portugal - OA2 - 0.51 - 0.84 -0.77 - 0.59 -0.55 -0.54 -0.59 - 0.55 - 0.53 - OA8 -0.68 - 0.75 - 0.69 - 0.69 - 0.62 Malta - 096 - 076 - 0.80 -0.79 - 0.73 - 0.62 - 073 - 077 - 0.65 - 0.67 -0.67 - 0.68 -073 - 0.91 -0.93 -1.02 Jordan 0.65 0.55 - 0.27 - 0.38 - OA1 -0.74 -1.09 - 0.93 - 1.72 -1.10 - 0.76 - 0.67 -1 .12 - 0.96 -0.82 -1 A7 Egypt -0.99 - 0.65 - 0.91 -1.57 - 1.24 0.13 -1.01 - 2.06 - 1.64 -2.16 -1.71 -1.68 - 1.16 -1AO -1 .65 -1.72 Libya -1.36 -1.38 -1.29 - 1A8 - 1.76 -1.66 -1.58 - 2.10 - 2A4 - 2.50 - 2.17 - 2.52 - 2.39 -2.76 - 2.67 - 2.58 Algeria -1 .66 - 1.74 -1 .61 - 2. 15 - 1.94 - 2.62 - 2.65 - 3A8 -2A9 - 2.85 - 2.87 - 2.90 - 3.54 - 3.20 - 3A8 -3.86

Note: Countries are ranked by the RC index in 1997.

47

MEDIT N" 412001

Table 6 Ranking of Mediterranean countries according to their RXU and RTU indices.

Country RTU RXU CVXRSA CVRe RXU RTU Number of

Rank Rank Index Index products

Spain 12 90 276.6 22.9 0.79 0.07 348 Turkey 19 124 213.5 22.8 0.89 0.10 239 Italy 24 57 273.4 49.1 0.68 0.12 404 Malta 26 108 29.7 4.9 0.85 0.14 35 Algeria 27 151 15.3 2.2 1.02 0.15 15 Portugal 54 145 255.2 96.9 0.98 0.37 261 Jordan 60 159 125.6 47.5 1.06 0.40 119 Morocco 80 135 98.6 56.6 0.92 0.53 107 Tunisia 91 130 73.7 47.6 0.91 0.59 81 Cyprus 127 131 145.6 149.5 0.91 0.93 160 Lebanon 129 32 28.5 48.5 0.56 0.95 51 Egypt 149 156 141.8 207.9 1.03 1.51 138 France 151 21 208.7 689.2 0.49 1.63 423 Greece 162 138 225.5 679.7 0.93 2.81 242 Syria 164 170 93.9 270.1 1.12 3.22 84

4r-------------~--------------~

3

X R 2 C A

Italy d=l

Portugal

I I I Greece I 0 I I I I I I I Spain I 0 I France I 0

I I I I I I I I

o~ ______________ ~ ______________ ~ -2 0 2

RC

Figure 4 - Plotting EU Mediterranean countries according to their indices of RXCA and RCfor Total of Agricultural Products in 1997.

CA and RC (CVXRCA and CVRC) for all products exported, which used in calculating RXU and RTU respectively. As shown by Table 6, the rank obtained using uncer- . tainty indicators contrast previous rank based on XRCA and RC indices. Particularly, Spain was the Mediter­ranean country with the less uncertain agricultural trade given its low RTU value (0.07), a value that placed it in the 12th world position. Turkey and Italy followed with RTU values of 0.10 and 0.12 respectively. France, fol­lowed by Italy, showed the less uncertain agricultural exports with values of 0.49 and 0.68 respectively. In sharp contrast to XRCA and RC values, Greece exhibit­ed high values of both export and trade uncertainty el­ements. Figure 5 plots the Mediterranean countries ac­cording to their world ranking on RXU and RTU in-

48

s- 0.8 >< !:5.

Jordan 0

EgYRt Syria Algeria 0 0

0 Portugal Greece 0 Morocco Cyprus 0

Turkey 0 0 0 Tunisia

i!:' I: .\; 0.6 'I: ., co I: :::0

Malta 0 0

Spain lwoRLo 0 (0.5,0.2)

'I: Q 0.4 Cl. ..

W Italy .., ., 0 'ii ., > 0.2 CD a:

Lebanon 0 France

0

0 o 0.2 0.4 0.6 0.8

Revealed Trade Uncertainty (RTU)

Figure 5 - Mappillg Revealed Agricultural Trade Uncertaillty q/Medilerraneall Countries.

dices. As can be seen by Figure 5, which also plots world uncertainty, Italy appeared to be in most secure position in terms of export and trade uncertainty, fol­lowed by Spain which has a little more volatile advan­tage in exports. France, although it presented the least volatile export advantage among Mediterranean coun­tries, its high RTU value indicates instability in trade of agricultural products, which can be attributed to volatile imports.

DISCUSSION

Trade and export performance have been used as plau­sible indicators of competitiveness. Balassa (1965) was the first to develop the revealed comparative advantage index (RCA), which can be a proxy of a country's com­petitiveness. However, the initial export-performance formula (1) suffers from methodological limitations, the Revealed Symmetric Comparative Advantage index (RSCA) was developed as a better indicator of compar­ative advantage. Vollrath and Vo (1988) developed the Revealed Competitiveness index (RC), which is similar to GL index of intra-industry trade, incorporating equal­ly import and export figures. In this way, the RC index divulges a country's trade pattern better than indices ex­clusively based on exports. Nevertheless, Both RCA and RC indices suffer from threats to their validity and relia­bility stemming from the distortions in trade as imposed from governmental interventions, protectionism poli­cies, etc. Uncertainty can be considered an alternative proxy of comparative advantage since, as every policymaker would tell, it generates detrimental predictions of a country's course of trade. The long-term variability of revealed comparative advantage was used to develop indices of international trade uncertainty. Two such in­dices were developed: (a) the revealed Export Uncer-

MEDIT N° 412001

tainty (RXU) and (b) the revealed Export Uncertainty (RTU). RXU was based on the CV of RSCA index, thus it was assumed to indicate export uncertainty and RTU was based on the CV of RC index, indicating trade un­certainty. The values of XRCA and RC indices of total agricultural products present Greece the most competitive country in the Mediterranean basin, with the rest EU states to lag considerably behind. France and Italy appear to be less competitive than Greece even Cyprus. On the other hand, Spain and Portugal seem to have export disad­vantage and comparative disadvantage in trading agri­cultural products a performance comparable to what Morocco, Malta, Tunisia, and Jordan have achieved with far less export volume. Uncertainty indices give a more anticipated picture. Par­ticularly, based on the RXU and RTU values, it is demonstrated that Italy has developed a stable compar­ative advantage which is revealed by low uncertainty terms, lower than world standards. In sharp contrast, Greek trade face considerable uncertainty in both ex­port and overall trade terms. As a consequence, Greece, along with Syria, Egypt, and Cyprus which display sim­ilar RTU and RXU values, occupy the less favoured po­sitions in the international trade of agricultural products amongst Mediterranean basin states. Certain limitations constrain the interpretability of the RXU and RTU indicators. Particularly, XRCA and RC in­dices may produce unreliable results to the extent that trade distortions are imposed. However, the computa­tion of uncertainty indicators over a significant number of products relaxes this reliability bias. Furthermore, it can be argued that RXU and RTU indices reveal the im­pact of policy intervention as one of policy objectives it to reduce uncertainty. Data availability and validity also restricts the interpretability of RXU and RTU indices. Further research is needed in order to examine whether indicators of uncertainty could be developed using oth­er methods like GARC and/or using other indicators of comparative advantage. Moreover, the indices of uncer­tainty could be tested as predictors of trade perfor­mance since uncertainty is assumed to irreverSibly re­late to comparative advantage. •

REFERENCES Abbasi, (AI) M.A. (991), The Effect of Oil Revenues Instability on the Oil Based and Labour Exporting Economies: The Case of the Arab Region, Doc­toral Dissertation, University Of Missouri, Columbia.

Anderson, M.A. (990), The Formation And Effects Of Tariff Preference Schemes, Doctoral Dissertation, The University of Wisconsin, Madison.

Balassa Bela (1965), Trade Liberalization and Revealed Comparative Advan­tage, The Manchester School, 33 (2), 327-345.

Edison, H. and Melvin, M. (990), "The Determinants and Implications of the Choice of an Exchange Rate System." In: W. Haraf and T. Willet, editors. Monetary Policy for a Volatile Global Economy. Washington, D.e., United States: AEI Press.

European Commission (1994), An Industrial Competitiveness Policy for the

49

European Union, Bulletin of the European Union, Luxembourg: OOPEe.

Gallagher, P. (998), International Marketing Margins for Agricultural Prod­ucts: Effects of Some non-tariff Trade Barriers, American Journal of Agricul­tural Economics, 80 (2), 325-336.

Ghirmay, T. (997), Outward Orientation And Economic Growth In Less De­veloped Countries: An Empirical Investigation, Doctoral Dissertation, South­ern Illinois UniverSity, Carbondale.

Goldin, I. (1990), Comparative Advantage: Theory and Application to Devel­oping Country Agriculture, OECD Developing Centre, Technical Papers, 16, June.

Haberler, G. (1977), Survey of the Circumstances Affecting the Location of Production. In B. Ohlin (eds.) The International Allocation of Economic Ac­tivity: Proceedings of a Nobel Symposium, Holmes and Meier, New York.

Haley Stephen, L. (1987), Conceptual Model of Competitiveness and Com­parative Advantage in Agricultural Trade, U.S. Department of Agriculture, Economic Research Service, Agricultural and Trade Analysis Division, ERS Staff Report No. AGES870513 Quly).

Haque I. (1995), Trade, Technology, and International Competitiveness, EDI Development Studies, Economic Development Institute of the World Bank, December.

Ibarra, P.L.A. (1992), Credibility Of Trade Policy Reform: The Mexican Expe­rience, Doctoral Dissertation, University of California, Los Angeles.

Kilkenny, M. and Nalbarte, L. (2000), Keystone Sector Identification: A Graph Theory-Social Network Analysis Approach, Regional Research Institute, WVU (http://www.rri.wvu.edulWebBooklKilkenny/editedkeystone.htm).

Laursen, K. (1998), Revealed Comparative Advantage and the Alternatives as Measures of International Specialisation, Working Paper No. 98-30, DRUID Danish Research Unit For Industrial Dynamics, December.

Marrewijk, van e., Bergeijk, van P.A.G. (993), Endogenous Trade Uncer­tainty: Why Countries May Specialize against Comparative Advantage, Re­gional Science and Urban Economics, 23 (5), 681-694.

Palkhiwala, J.N. (1993), Export Instability In Less Developed Countries, Doc­toral Dissertation, California State University, Fullerton.

Perkins, P.R. (1987), "Measuring Economic Competitiveness in Trade". In: U.S. Competitiveness in the World Wheat Market: Proceedings of a Research Conference, U.S. Department of Agriculture, Economic Research Service, In­ternational Economics DiviSion, ERS Staff Report No. AGES860903 (March).

Ricardo, D. (1817), On the Principles of Political Economy and Taxation, John Murray, London.

Smets, F.R. (1993), Essays On Foreign Direct Investment, Doctoral Disserta­tion, Yale University.

Stilwell, F. (2000). Dollar Down: Uncertainty Up, Journal of Australian Polit­ical Economy, No. 46, December, pp. 38-43.

Traill, B. and Silva (da), J,G. (1996), Measuring International Competitive­ness: the Case of the European Food Industry, International Business Re­view, 5 (2), 151-166.

Vollrath, T.L. (1987), Revealed Competitive Advantage for Wheat. In: U.S. Competitiveness in the World Wheat Market: Proceedings of a Research Con­ference, U.s. Department of Agriculture, Economic Research Service, Inter­national Economics DiviSion, ERS Staff Report No. AGES860903 (March).

Vollrath, T.L. and De Huu, Vo. (1988), Investigating the Nature of World Agri­cultural Competitiveness, U.S. Department of Agriculture, Economic Re­search Service, Technical Bulletin No. 1754 (December).

Wei, 5.]. (997), Why is Corruption So Much More Taxing Than Tax? Arbi­trariness Kills, National Bureau of Economic Research, Working Paper: 6255, November, pages 17.

Yanagida, J,P. and Tian, X. (996), Competitiveness and comparative trade advantage: an empirical analysis of selected pacific basin and Asian coun­tries, Asia Pacific Food and Agric. Trade Working Group (APFAWG-L).