An Exchange Rate System to Facilitate Trade Cooperation in the Agricultural Sector Between China, Japan, and Korea

In this paper, we propose the use of a system of exchange rates for the agricultural sector (ERAs) based on market-determined exchange rates adjusted for changes in real eff[r]

International Trade Department, Chungnam National University, Korea

■2012 JSPS Asian CORE Program, Nagoya University and VNU University of Economics and Business

An Exchange Rate System to Facilitate Trade Cooperation in the

Agricultural Sector Between China, Japan, and Korea

Chungnam National University chan-guk HUH*

ABSTRACT :The problem of how to broaden access to one another‘s markets in the agricultural sector remains the key

obstacle to establishing an FTA among China, Japan, and Korea (CJK) This paper takes the Common Agricultural Policy

(CAP) of the EU as a reference model for future CJK agricultural cooperation and offers suggestions on exchange rate

issues in the context of prospective trilateral trade and policy coordination in the agricultural sector This paper finds CJK

trade in the agricultural sector to be negligible in magnitude compared to that in the manufacturing sector, but not entirely

inelastic to exchange rates and income growth Expansion of agricultural trade is expected to open a new dimension of

benefit by moderating the adverse welfare effect of high food prices in Japan and Korea, where the proportions of the

population in income groups with high Engel‘s coefficients are increasing

KEYWORDS : China–Japan–Korea trade, agriculture, real effective exchange rate, CAP, Engel‘s coefficient

1 Introduction

Economic interactions among countries in East Asia

have grown by leaps and bounds in recent decades

The emergence of a rapidly growing China has

quickened the pace of regional economic exchange

Japan and Korea have established a trade network also

including China, which has expanded especially

rapidly since 2001, when China joined the World Trade

Organization These three countries—China, Japan and

Korea (CJK)—have individually pursued active

regional trade agreements (RTAs) with many partners

near and far; however, they still do not have a RTA

among themselves A key obstacle has been

intransigence on the part of Japan and Korea about

opening their agricultural markets (Urata, 2007;

Mulgan, 2008; Kuno and Kimura, 2008; Anderson,

2009; Huh, 2012a) This oddity is hard to justify for at

least two reasons First, Japan and Korea could

potentially gain a large amount of purchasing power by

lowering their food costs, and thus lower the cost of

living for their largely urban populations if they were to

source more agricultural goods from lower-cost

regional suppliers, that is, Japan from Korea and China,

and Korea from China Second, CJK are at the core of

current discussions to deepen East Asian economic

integration in areas such as capital markets and

cooperative exchange rate arrangements.1 Building

mutual trust through repeated interactions with

potential participants is an essential ingredient of making progress in such an endeavor Thus, if possible, establishing a workable trilateral arrangement to boost trade among CJK in agriculture, which lags far behind other industries in this regard, will be an ideal facilitator for trust-building among these countries

This paper examines the current state of CJK agricultural trade with special attention to issues related

to exchange rates We use exchange rates between CJK currencies in two different contexts: to retrospectively assess the responsiveness of regional agricultural trade

to exchange rate changes, and to discuss prospective exchange rate arrangements to boost CJK cooperation and trade in agricultural sectors down the road

Many papers examined various aspects of CJK trade

in conjunction with a hypothetical ―CJK FTA,‖ but only few have focused on trade in agricultural sectors.2 Intra-CJK trade has mostly been in manufacturing sectors, in which trade patterns have been undergirded

by the major non-regional trading partners—the United States and the European Union—which are the main markets for the finished products.3 In contrast, trade in agriculture in CJK can be expected to be inwardly oriented, as the products are mostly consumed within the importing countries Accordingly, patterns in agricultural trade can be expected to be different from those of trade in manufactured goods Our examination will shed some light on the sector-specific differences

in intra-CJK trade patterns, such as their

responsiveness to typical explanatory variables of

exchange rate and income growth

In terms of the prospective role of exchange rates in

CJK cooperation in the trade of agricultural products,

the experience of Europe in the early period of the

Common Agricultural Policy (CAP) offers a useful

reference In early days of the CAP, coordination of

inter-country subsidies and transfers required a unit of

account of commodity values—which is inextricable

from the exchange rates of member countries (Josling,

1969; Cramon-Taubadel, 1994) Exchange rates had

not presented any difficulties initially, in the 1960s, as

the currencies of the participating countries were

securely linked by the fixed exchange rate regime of

the Bretton Woods system The collapse of this system

in the early 1970s and the exchange rate instability

observed in ensuing decades in European currencies

hugely complicated the operation of the CAP

In a strong contrast with the European situation,

neither free trade of farm goods nor exchange rate

cooperation yet exists among CJK Given that discord

about agricultural trade has been a major block to past

attempts at a CJK RTA, progress in this area will likely

be a prerequisite to achieving a meaningful RTA

between these countries Successful scaling of the

agriculture hurdle will most likely involve trilateral

cooperation in terms of price coordination as well as

some scheme of cross-border financial transfers similar

to the type seen in the early CAP period The efficient

operation of such mechanisms would necessarily

involve discussion of how to deal with exchange rate

volatility, as a currency-cooperation arrangement that

ties trilateral exchange rates together closely is not

likely to emerge in the near future In this paper, we

propose the use of a system of exchange rates for the

agricultural sector (ERAs) based on market-determined

exchange rates adjusted for changes in real effective

exchange rates for the agricultural sector (REERAs) as

a means of surmounting the above-delineated issues in

the coordination of CJK agricultural trade in the future

The use of market-determined exchange rates would

lower the likelihood of complications in cross-border

agricultural coordination the CAP experienced due to

the use of an inflexible unit of accounts The

incorporation of REERAs, which reflects factors

affecting an individual country‘s competitiveness as

well as actual sector-specific trade patterns, reflects the

fact that ERAs are adopted as the unit of account to be

used for trade in agricultural goods

The remainder of the paper is organized as follows

Section II describes the potential benefits of lower food

costs for Japan and Korea Section III examines

intra-CJK trade patterns in agricultural as well as

non-agricultural sectors Section IV discusses how

exchange rate issues have complicated coordination

among CAP member countries in the past In Section V, REERAs are introduced, and how they can be applied

to CJK agricultural trade is illustrated Section VI concludes

II Expected benefits of lower food costs for Japan and Korea

Previous studies that have examined the possibility of

a CJK FTA have almost without any exception suggested that China and its agricultural sector would gain in such an arrangement (TJR, 2004; Yoon et al., 2009) Thus, in this section we limit our focus to the other two countries involved, Japan and Korea We will consider three reasons why enhanced inter-CJK trade

in agricultural goods could bring about improvement in welfare First, it is a widely recognized fact that the costs of many farm goods are higher in Japan and Korea (JK) than in most other countries but especially compared to China A typical ranking of prices for most agricultural goods would be Japan, Korea, China,

in descending order.4 Average agricultural protection levels measured in terms of the nominal rate of protection (NRP) are high in JK—the NRP measures the difference between domestic and international prices (Honma, 2011) Given such a situation, it does not take much imagination to recognize that lowering protection levels in JK and allowing freer trade in farm goods within CJK would have a moderating effect on domestic food prices in the two high-income countries The second reason is that consumers in JK appear to spend a disproportionately high share of their income

on food, as can be seen in the international comparison

of household spending on food in <Table 1> Japanese consumers in particular spend a higher proportion of their budget on food compared to their counterparts in other high-income countries For example, an average Japanese household spends about the same overall amount per year (close to USD 23,000) on goods and services as those in Germany and Sweden However, where food spending is concerned, German and Swedish households respectively spend approximately USD 900 and USD 550 less on food than their Japanese counterparts In general, economies that rely more on food or agricultural imports tend to spend less

on food than similar-income countries The last column

of <Table 1> demonstrates this general pattern (note the exception of the US) For example, in the high-income group, consumers in both Singapore and Hong Kong, which do not have any indigenous agriculture and thus rely almost completely on agricultural imports spend proportionately far less on food than consumers in the East Asian high-income countries of Japan and Korea Such a pattern does not seem limited to high income countries Of the Association of Southeast Asian Nations countries

(ASEAN), Malaysia, whose level of import of

agricultural goods is the second-highest (next to

Singapore) among ASEAN members, has a

remarkably low share of spending dedicated to food for

a middle-income nation (even lower than the high-income nations of Japan and Korea)

<Table 1> International comparison of food-at-home budget share (average of 2008-2010, %, USD)1

Country

Share of household final consumption (%)

Total household final consumption expenditure (USD)

Expenditure per capita on food (USD)

Agricultural imports per capita3 (USD) (B)/(A) ×100

(%)

1 Simple averages of data for three years (2008–2010) are used to minimize the effects of fluctuations in the exchange rates of these countries vis-à-vis the US dollar in this period All data except those on imports per capita are from the United States Department of Agriculture website

2 Alcoholic beverages and tobacco

3 Agricultural import data are for SITC 0 imports, from the UN‘s Comtrade dataset Populations (for 2011) were taken from the CIA website Averages of the three years 2008, 2009, and 2010 were divided by population to obtain ―agricultural imports per capita.‖

Sources: USDA (http://www.ers.usda.gov/data-products/food-expenditures.aspx , accessed on July 30, 2012); CIA World Factbook

( https://www.cia.gov/library/publications/the-world-factbook/rankorder/2119rank.html , accessed on August 4, 2012)

Third, the issue of food prices has become

increasingly significant in JK for two related reasons:

aging populations and growing income inequality

Rapid aging (along with the problem of two-tiered

employment, with regular vs non-regular workers in

the labor market) has contributed to rising income

inequality in Japan (Jones, 2007) The Korean

population is also rapidly aging This trend has been

accompanied by a noticeable rise in poverty rates

among the senior population.5 For the low-income

group, paying for food takes up a bigger share of their

limited budget; or, equivalently, their Engel‘s

coefficients tend to rise Rise in household income

inequality in JK has put the issue of costs of living for

different income groups at the forefront of policy

deliberations.6 Urakawa and Oshio (2010) motivate

their study on commodity tax reform in Japan and

Korea as follows: ―… there is a possibility that a Value

Added Tax (VAT) increase will be considered as a

plausible policy option in both countries, because

demographic pressures due to population aging will continue to put pressure on government spending over the coming decades‖ (p 580) They examine the impact of the marginal costs of taxing major commodity groups (e.g., food, education, transportation, etc), seeking to identify a good with a higher (lower) marginal cost of tax, whose tax rate should be lowered (raised) to improve welfare One of their findings is ―… revenue-neutral marginal tax reforms incorporating a reduced tax on food & beverages are more likely to face an efficiency-equity tradeoff in Korea than Japan This reflects more uneven spending on non-necessities across income classes in Korea, causing more income redistribution through tax rate changes.‖ (p 592)

An alternative way to reduce food costs, as opposed

to lowering VAT, would be lowering NPRs on key food items This would not be a clear-cut Pareto improvement, as it implies some negative income consequences for some farm sectors However, it

should be possible to design a commensurate direct

income subsidy scheme for the affected farmers Either

option might mean the same amount of financial

resources, but NPR reduction would involve less

potentially distorting effects on household choices

This would also have the sanguine effect of lowering

the proportion of consumer-financed (versus

taxpayer-financed) agricultural support in JK, which

are currently quite high when compared to those in

other OECD countries (OECD, 2010).7 In short, the

examinations in this section point to the possibility that

increased trade in agricultural goods would bring about

welfare improvements to a large number of households

in both Japan and Korea

III Intra-CJK trade patterns in the agriculture

and manufacturing sectors

We examine intra-CJK trade patterns of the

agriculture and manufacturing sectors, whose figures for selected years are shown in <Table 2> Contrasts in the trends for regional exports in agriculture (SITC 0) and manufacturing (SITC 5–8) are quite striking For China, agricultural exports to JK grew 80% and manufacturing exports over 300% between 2000 and

2010 The ratio between these Chinese exports of agricultural and manufactured commodities grew from 1:6.74 in 2000 to 1:15.2 in 2010 For Korea, the divergence in the two sectors‘ export performance over the same period is even starker; the ratio rose from 1:18.3 in 2000 to 1:65.4 in 2010 As agricultural exports grew a mere 17.8%, manufacturing exports surged 320% in the same period Similar patterns also can be seen for Japan Shares of agricultural exports out of CJK total exports shown for 2010 are notably smaller compared to the value-added shares of GDP accounted for by the agriculture sector

<Table 2> Intra-CJK exports in agriculture and manufactured goods

Year

Agriculture (SITC 0), in billion USD Manufacturing (SITC 5–8), in billion

USD China

to JK

Japan

to CK

Korea

to CJ

China

to JK

Japan

to CK

Korea

to CJ

Share of total intra-CJK exports accounted for by each country in 2010 (%)

Value added to GDP in 2010 (%)1

Source: UN Comtrade database

1 These are from the World Bank database The numbers for manufacturing are as defined in this database, and are broader than

manufacturing as defined in SITC 5–8 ( http://data.worldbank.org/indicator , accessed on Aug 14, 2012)

<Figure 1> Intra-CJK manufacturing and agricultural export growth rates (1997–2010, year on year)*

*Each line represents the annual growth of the USD value of one country‘s exports to the other two of CJK Manufacturing data includes SITC 5–8, and agriculture includes SITC 0 Source: Comtrade

The annual growth rates of the manufacturing and

agricultural exports of each CJK country are compared

in <Figure 1> For the manufacturing sector, patterns in

export growth rates among CJK are closely

synchronized and show no discernible differences;

contractions occurred in 1998 and 2009, both

associated with major regional or global economic

crises, with one additional dip in 2001 for Japan and

Korea This close co-movement reflects the vertical production fragmentation networks in effect across CJK (Kimura and Obashi, 2011) In a strong contrast to the manufacturing situation, patterns in CJK agricultural exports show a remarkable lack of coherence Thus, the two sectors differ from each other

in terms of magnitude as well as of their dynamic co-movement patterns

<Table 3> Error correction model (ECM) of sector-specific exports1 (1998.Q1–2010.Q4)

∆xt = a 0 + a 1∆xt-1 i + b l ∆re t-l i + c l ∆y t-l j + dv t-1 + et,

where v t-1 = x t-1 – αre t-1 – βy t-1 and e t ~ N(o, σ)

i = China

Manufacturing exports

Agricultural exports

i = Japan

Manufacturing exports

Agricultural exports

i = Korea

Manufacturing exports

Agricultural exports

1 See footnote 8 for data description Log-level series were tested for unit roots using the ADF test All series were found to have unit roots, and subsequently, cointegration tests were conducted for each export equation using the JJ test.Table 3 here

Next, we turn to a more systematic time series

regression framework to examine the connection

between sector-specific exports (in manufacturing and

agriculture, respectively) on the one hand and real

sector-specific exchange rates and income growth on

the other by estimating an export equation All

variables used for the analysis are quarterly series

measured in real terms Agriculture and manufacturing

exports are deflated with reference to the available

corresponding unit export price series Sector-specific

real exchange rates are calculated by deflating nominal

exchange rates using sector-specific producer price

indexes (PPIs) for the manufacturing and agricultural

sectors, obtained from national sources.8 For example,

the real exchange rate of the Korean won vis-à-vis the Chinese yuan is

Here, E cdenotes the nominal exchange rate of the Korean won to the Chinese yuan (the number of won per yuan) Pk and Pc each denote the producer price indexes (both manufacturing and agricultural) for Korea and China

Each series was tested and all were found to have unit roots using the augmented Dickey–Fuller (1981) test Subsequently, the cointegration test by the Johansen and Juselius (1990) was applied to each set of variables

that appear in the same export equation (e.g., levels of

real exports, real bilateral exchange rate, and importing

country‘s real GDP) Cointegrations indicating a

long-term relationship among a group of variables

were found for six out of the twelve cases Finally,

error correction models for those six cases, with

cointegration, and regressions using first-differenced

series were estimated Included in each equation were a

lagged dependent variable, a derived error term, lagged

once, from the corresponding cointegration equation (if

it existed), and three lags of explanation Lag length for

explanatory variables were adjusted to two or three,

depending on whether there was any significant entry

in equation Results are shown in <Table 3> Each

country block in the table has two sub-groups for

manufacturing, and agricultural exports, and in turn

each industry group contains two export destinations

Each cell either shows a statistically significant entry as

well as the significance levels or, for cells with ―not

sig.‖ and ―no cointegration‖ respectively, denotes cases

where there are no significant explanatory variables

and no cointegration The divergence in the regression

results between the two sectors is noticeable Results

show cointegrated relationships between

manufacturing exports, real bilateral exchange rate, and

real GDP of importing country in four out of six cases

In comparison, that is true for only two out of six

agricultural export equations In the manufacturing

group, Japan‘s exports to China are the only case

showing no significant short- or long-term relationship

between variables, while in the agricultural export

group no significant short or long-term relationship

was detected in any of the three cases (China‘s exports

to Korea, Japan‘s exports to China, or Korea‘s exports

to China)

These results for the agricultural sector seem to confirm that there is a dearth of intra-CJK trade in agricultural goods, but at the same time show a responsiveness of agricultural trade flows to economic factors, albeit a weak one The latter finding is indeed encouraging, as it indicates the possibility for further expansion of agricultural trade The challenge appears

to be finding ways to lower existing agricultural trade barriers so that CJK consumers benefit in terms of price as well as quality from access to a wider range of farm goods Given the high levels of border protection and domestic agricultural subsidies seen in CJK, closer coordination of agricultural policies by the CJK governments should facilitate a freer flow of agricultural goods across CJK To achieve this, the European experience of the CAP, which we turn to next, can offer useful hints

IV CAP, an example of cross-country coordination for agricultural trade among CJK

We start by looking at the differences between the situations in CJK and Europe with regard to intra-regional trade patterns in the agricultural and manufacturing sectors Patterns of exports in the two sectors among three key European countries—France, Germany, and Italy—in the 1980s can be seen in

<Table 4>.9 Compared to the CJK patterns in <Table 2>, the European countries‘ trade in the two sectors was much more balanced in terms of absolute size as well as relative to its shares in the respective economies

<Table 4> Intra–France–Germany–Italy exports of agriculture and manufactured goods

Year

Agriculture (SITC 0), in billion USD Manufacturing (SITC 5–8), in billion USD France

to GI

Germany

to FI

Italy

to FG

France

to GI

Germany

to FI

Italy

to FG

Share of total ―FGI‖ exports accounted for by each country in 1990 (%)

Value added to GDP in 1990 (%)1

Source: Comtrade

1 These are from the World Bank database The numbers for manufacturing are as defined in this database, and are broader than

manufacturing as defined in SITC 5–8 ( http://data.worldbank.org/indicator , accessed on Aug 14, 2012)

Among factors that could explain the different

situations in CJK and Europe the fact that European

countries had been members of the customs union,

whereas CJK have imposed high tariffs, ranging from

approximately 12% (Japan on imports from Korea) to

140% (Korea on imports from China), on agricultural

imports from each other stands out.10 Furthermore, the

early members of the customs union took active measures to facilitate smooth flow of agricultural goods across countries, namely the CAP.11 By design, agricultural support systems in European countries were based on the principle of supporting market prices With the advent of the CAP, prices in different countries needed to be compared and harmonized for

free trade of farm goods The price support system has

several key numerical indicators, including target price,

intervention price, and price floor for a product (Josling,

1969) Intra–customs union trade of farm goods in

Europe also entailed levies and subsidies.12 Thus, the

fact that prices were measured in different currencies

made exchange rates between CAP member countries

a problem that was innate to the operations of the CAP

Exchange rates were not a particularly important issue

for CAP members in the early days, as cross-country

harmonization was done under fixed exchange rates

between member currencies.13 However, volatile

exchange rates became a reality for CAP countries with

the breakdown of the Bretton Woods system This, in

turn, disrupted a finely calibrated regime of

multi-country harmonization requiring cumbersome

adjustments; in particular, compensatory payments

became necessary, as members wanted to shield their

domestic agricultural markets and prices from large

changes in exchange rates Price differentials existed

among member countries despite designated common

prices for farm goods, and each country tried to

maintain its own domestic prices while allowing free

intra-regional flow of trade in farm goods at the same

time This required a system of payments known as

Monetary Compensatory Amounts (MCA):

―In the second half of 1969, the FF and the

DM were de- and revalued 11.11% and 9.29%

respectively Since the CAP‘s unit of account

was a non-member currency, these changes

should have led … to corresponding increase

(France) and decrease (Germany) in CAP

prices The French and German governments

were unwilling to accept these consequences

and obtained permission to continue using

their old pre-change parities as ―green rates‖

for agricultural purposes Of course, at market

exchange rates, German agricultural prices

were now higher than in the rest of the EC, and

French prices were lower To prevent the trade

flows that these price differences would

otherwise have engendered the first Monetary

Compensation Amounts (MCA) taxes and

subsidies on agricultural trade were created as

part of what was thought to be a temporary

arrangement …‖ (Cramon-Taubadel, 1994, p

113)

The circumstances with CJK are quite different: there

is no comparable mechanism to the European CAP,

and intra-regional trade in agricultural goods is

insignificant, as shown in Section III and in a

comparison between <Table 2> and <Table 4>

However, there is a growing realization that regional

cooperation in agricultural areas that is roughly

modeled after the CAP is crucial for further deepening

of economic cooperation in East Asia (Honma, 2006, 2011), and Northeast Asia (Suzuki et al 2007, Huh, 2012a) Furthermore, due to purely domestic considerations, food costs are likely to gain more economic and political importance in Japan and Korea

in the future, as discussed in Section II This, in turn, raises the likelihood of the realization of a formal CJK cooperation arrangement in agricultural areas When this happens, the issue of the exchange rate that is to be used will emerge as a central topic That is because there are large price gaps across CJK for many commodities, which will require some scheme to reduce price differentials, probably involving modalities similar to ones seen in the early period of the CAP Both the harmonization of price differentials and the pooling of funds will necessarily involve converting numbers enumerated in different national currencies into a common unit One important lesson

of the MCA experiences of the early CAP for us in the post–Bretton Woods era is that it is better to minimize the fixity of the exchange rates used for international coordination when a durable anchor is absent We turn

in the next section to CJK exchange rate issues, keeping this lesson in mind

V Exchange rates and agricultural sector real effective exchange rate issues for CJK

1 Definitions

Exchange rates are going to matter very much at the initial stage of agricultural coordination for CJK Comparing prices of agricultural commodities among CJK will necessarily involve exchange rates, as will the subsequent settling of balances relating to cross-country financial obligations arising from whatever cooperative mechanisms adopted Using ordinary market-determined headline exchange rates is one option However, excess volatility in exchange rates, seen around 1997 and 2008 for example, could have detrimental effects on nascent intra-CJK agricultural trade based on studies that document negative effects of exchange rate volatility on trade.14 Thus, a unit of account, to be referred as the exchange rate for agriculture (ERA), is needed A candidate for this unit should be somewhat removed from market exchange rates, which will dampen price fluctuations One possibility would be to use moving averages of market-determined exchange rates for the past four quarters This would reduce ERA volatilities compared

to the use of real-time spot exchange rates However, situations might arise calling for discrete judgmental adjustments in ERAs For example, major idiosyncratic changes in factors other than exchange rates that affect agricultural sector of a participating country

specifically such as production costs would necessitate

discrete changes in exchange rates used for

cross-country coordination For this we turn to a

method based on the real effective exchange rate of

agricultural sector (REERA), which is a variant of the

real effective exchange rate (REER) widely used in

international economics (Chinn, 2006; Goldberg, 2004;

Klau and Fung, 2006) REERs for many countries are

regularly updated by, for example, the US Federal

Reserve Board and the Bank for International

Settlement REERA is a useful measure, as it reflects

various factors related to the external competitiveness

of a country‘s agricultural sector: nominal exchange

rates, costs of agricultural outputs, and agricultural

trade value REERA can be used as follows Once an

ERA is chosen, it will be used until the date of the next

adjustment agreed to by CJK When resetting the ERA,

changes in REERA will be taken into consideration in

determining the amount of the adjustment

The REER of country i vis-à-vis a set of trading

partners indexed by j is a weighted average of

corresponding bilateral real exchange rates Let us use

the Korean REER vis-à-vis China and Japan as an

example in the following exposition It is derived as the

product of bilateral real exchange rates with geometric

weights as follows:

Here, and respectively denote the bilateral

real exchange rate of the Korean won vis-à-vis the

Chinese yuan and the Japanese yen, as shown in (1),

and wc and wJ denote the weights of China and Japan

(e.g., wc + wJ = 1) in the real effective exchange rate of

the Korean REER; they are determined as follows:15

. (3)

In equation (3), Xk and Mk each denote the values of total exports and total imports This is multiplied by the second term, which is the ratio of Korea‘s exports to China over the sum of Korea‘s exports to China and Japan The structure of the second term is same as the first term, with exports replaced by imports The weight of Japan in (2), wJ, is calculated in the same way We convert the real exchange rates into index form by dividing each of the series by the 2005 value Thus, the real exchange rates have been standardized

so that all REER values in 2005 will be 100 Since effective exchange rates are denoted in the ―direct‖ way, that is, by the amount of home currency units per one unit of foreign currency, an increase in the REER

of country i would indicate a gain in competiveness for

the exports of that country

Three REERs are derived for each of the CJK countries, two sector-specific exchange rates (one each for manufacturing and agriculture sectors) and an aggregate for all industries, respectively denoted REERM, REERA, and REER For sector-specific REERs, export and import data as well as price indexes

of the corresponding industry groups are used

REERs calculated using annual data from 1996 to

2010 are shown in <Figure 2>.16 The REER of China has remained close to 100 for the whole sample period, though it rose to approximately 107 in 2010 from approximately 92 in 2008 The Japanese REER has exhibited more movement: starting from approximately 90 in the late 1990s, it rose steadily to

110 in 2007 and then fell to 84 in 2010 The Korean REER also shows a distinct shift around 2007; it fell steadily from the 130 level in 1998 to the mid-90s in

2007 before rebounding sharply to the 120 level in

2010 For the recent (post-2007) period, the REERs of CJK have shown divergent movements

<Table 5> Producer Price Index for agricultural sectors and nominal exchange rates among CJK

<Figure 2> Trends in three REERs for each of CJK (Annual, 2005=100)1

China

Japan

Korea

1 See footnote 16 for data sources

REERMs and REERAs of China and Japan have

moved in closer alignment with their respective

REERs compared to those of Korea over the sample

period They nevertheless show different patterns The

Korean REERM has risen the most, from 83 to 135

between 1996 and 2010, while the Chinese REERM

has fallen from 118 to 85 over the same period In

contrast, their Japanese counterpart has not moved

much from the 100 level Turning to REERAs, the Korean REERA exhibits by far the most divergent patterns out of the three cases The key cause is the rapid rise in agricultural prices in Korea over the sample period of 1996 to 2010, when Korean agricultural PPI rose from 72.6 to 120—a 65% increase, as shown in <Table 5> Chinese and Japanese agricultural PPIs rose 5% and -9.6%, respectively, over

the same period Nominally, the won has depreciated

quite substantially against both the yuan and the yen

Thus, the loss of agricultural competitiveness of

Korean agricultural exports measured in terms of

REERA over the sample period has been mainly due to rapid rise in production costs in the Korean agricultural sector

<Table 6> Bilateral exchange rates, moving averages, ERAs, and REERAs of China and Japan

Year

Nominal bilateral exchange rates and ERAs REERA JPY/CNY

(1)

Moving average

2 An illustration of the use of ERAs and REERAs

We will illustrate how the proposed concepts of ERA

and REERA can be implemented using the example of

China and Japan Let us assume that the hypothetical

CJK coordination discussed above had been going on

in the mid-2000s; <Table 6> shows quarterly data for

three bilateral nominal exchange rates; Japanese

yen/Chinese yuan (JPY/CNY); four-quarter (for

periods t-3, t-2, t-1, and t) moving average; and the

nominal exchange rate for 2005 Q1 to 2007 Q4; as

well as annual China–Japan (CJ) REERAs for the

same years The ERA column is for whatever exchange

rates CJ agree to use for agricultural trade cooperation

Suppose that CJK decided to use a four-quarter moving

average value of the third quarter each year, changed

once a year Then, the first ERA of 2005 for JPY/CNY

would be 13.02, and for 2006 and 2007, respectively,

14.47 and 15.41 These imply nominal depreciation of

JPY against CNY of 11.1% for 2006 and 6.5% for

2007 A simple method of updating ERAs

mechanically once a year in this case would mean large

consecutive changes in ERA values for China and

Japan

However, we propose a different approach, one that takes into account changes in competiveness in both countries measured in terms of REERAs The competitiveness of China‘s agricultural exports rose 2.8% between 2005 and 2006 but declined by 12.3% between 2006 and 2007, while Japanese REERA gained 4.1% and 15% consecutively in 2006 and 2007

So the net gain of Japanese over Chinese REERA is 1.3% point from 2005 to 2006, and in 2007, it is 27.3% point It seems sensible to consider this substantial divergence in competitiveness in setting ERAs Different methods could be used to take this issue into considerations to moderate its impact One way would

be to multiply the change in the moving average exchange rate by (1-λ), where λ is the net difference in the percentage changes the of REERAs of the two countries involved The formula for the ERA of period

t could be as follows: