The REER and mineral exporters : case study of Suriname

(1)

UNIVERSITY OF AMSTERDAM

FACULTY OF BUSINESS & ECONOMICS

The REER and Mineral Exporters:

Case study of Suriname

Thesis for the acquisition of the Master of Science Degree in Economics

Name: Mark Yngard Student number: 11575239 Supervisor: mw. Drs. Naomi Leefmans Second reader:Dr. D.J.M. Veestraeten

Amsterdam, July 13, 2018 Word count: 9206 words

(2)

Preface

This master thesis is written for the completion of the Master of Science in Economics program at the University of Amsterdam. Within the interesting area of International Economics & Globalization, I have gained much knowledge and also discovered an area where I like to specialize in the future.

Several persons have contributed academically, practically and with support to this master thesis. I would therefore firstly like to thank my supervisor Ms. Naomi Leefmans for sharing her knowledge, and especially in assisting with the quantitative part. At last, I want to thank everyone in my family, especially my father and mother for their love and support.

Statement of Originality

This document is written by Student Yngard Mark who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

From the end of 2015 to mid-2016, the official exchange rate of Suriname started to depreciate widely. Coupled with drops in gold and oil prices in the last 10 years, the main idea of this thesis is to ascertain the impact of mineral prices on the real effective exchange rate of Suriname.

An analysis is done to measure the impact of mineral prices on the real effective exchange rate of a panel set of countries with a time series of 1980 to 2011. Panel data are used to increase the number of observations and thereby the robustness of the analysis. Data from sources, such as the World Bank, Bruegel, IMF and the Central Bank of Suriname, are used to perform this analysis.

This thesis first calculated the real effective exchange rate for Suriname. To measure the impact of mineral prices on the real effective exchange rate of Suriname, an econometric model is built to perform a regression. The results reveal that shocks in mineral prices have a significant and positive impact on the real effective exchange rate.

(4)

List of Tables and Figures

Table 1: Overview of the Empirical Studies ... 14

Table 2: Final Set of Country-Commodity Pairs ... 24

Table 3: LLC Unit Root Tests for Mineral Prices ... 29

Table 4: IPS Unit Root Tests for Mineral Prices ... 29

Table 5: LLC Unit Root Tests for Real Effective Exchange Rates ... 30

Table 6: IPS Unit Root Tests for Real Effective Exchange Rates ... 30

Table 7: Co-integration Analysis ... 31

Table 8: Model 1 – DOLS Results ... 32

Table 9: Model 2 – FMOLS Results ... 33

Figure 1: Suriname, with Exchange Rates Trading Partners ... 18

Figure 2: The Real Effective Exchange Rate of Suriname ... 19

Figure 3: The Real Effective Exchange Rates of Panel Countries ... 20

Figure 4: Mineral Prices (2005=100) ... 22

(7)

6

Introduction

International trade is of great importance for the global economy, as economies benefit from international trade. Especially developing countries have experienced massive economic growth due to international trade. Suriname is also part of an integrated world and the country is a member of several trade unions. For a small country like Suriname in terms of capital and population, it is almost impossible to be self-sufficient. A challenge for this nation is to determine which goods should be traded to create maximum welfare. A report presented by the UNCTAD (2014) showed that the minerals oil, gold and aluminum contributed to about 90% of the export value in 2014. In periods when these mineral prices decline, it certainly puts pressure on the Surinamese dollar (SRD) through foreign exchange reserves deficits.

In recent years, the Central Bank of Suriname reformed its exchange rate policy towards a more flexible system. After this adjustment the nominal exchange rate started to depreciate strongly, at this point of time the exchange rate for US$ 1,- is SRD 7.5,-, while it was SRD 2.8,- eight years ago. This adjustment in the nominal exchange rate was necessary to decrease the deficit in the trade balance. Moreover, the last five years have witnessed by drops in oil and gold prices. From a theoretical perspective, such shocks in mineral prices might be affecting the real exchange rate by terms of trade and wealth effects. Precisely, this wealth effect that mineral prices may have on the real effective exchange rate is the topic of this thesis. The research question is as follows: ‘’ What was the impact of global mineral price volatility on the Real Effective Exchange Rate of Suriname in the period 1980-2011?’’

The Real Effective Exchange Rate (REER) is used in this thesis because it takes into account the changes in the competiveness of a country relative to its major trading partners. To my knowledge, there are no academic studies conducted to analyze the REER of Suriname. This research fills the gap by first calculating the REER for Suriname and then investigating the determinants of this variable, in particular mineral prices. As this study is the first attempt in this area, future studies can benefit from the findings. Moreover, the results could also prove to be useful for the evaluation of the domestic central bank and as advocacy for a future sovereign wealth fund.

To derive a conclusion, in order to increase the sample size, a panel data analysis will be used to investigate the effects of mineral price developments on the real effective exchange rates

(8)

7 of five countries, including Suriname. The other countries included are Chile, Guyana, Mauritania and, Peru. In addition, the model used is derived from former literature on this type of research. In particular, V. Bodart et al (2012), who estimate the effects of commodity price fluctuations on the real effective exchange rate for a panel of developing countries, formulate a simple relationship between the real effective exchange rate and the real price of the export mineral commodity of the country. Furthermore, V. Bodart et al (2012) state that, if other traditional determinants are relevant, the data is often not available or is of low quality for developing countries. Therefore, the simple model used by V. Bodart et al. will be used in this thesis.

This thesis consists of four chapters. Chapter one reviews the literature and empirical studies on the REER and the drivers of the real effective exchange rate. The second chapter deals with the selected variables of Suriname and the other countries regarding exchange rate and mineral price developments. Chapter three clarifies the choice of methodologies of data collection, regression and methods of data processing. The fourth chapter presents the data analysis and the results of the quantitative analysis. This thesis ends with conclusions and policy recommendations.

(9)

8

1. Literature Review

This chapter begins with a brief analysis of the concept of exchange rate in section 1.1.1 and of the real effective exchange rate in section 1.1.2. Thereafter, different theoretical approaches regarding drivers of the real effective exchange rate are reviewed in section 1.1.3, followed by a review of theory relating to the real effective exchange rate in relationship with mineral prices in section 1.1.4. Lastly, the impact of mineral prices on the real effective exchange rate in several empirical studies is explained in section 1.2.

1.1 Theoretical Review

1.1.1 The Exchange Rate

We first start with a brief overview of the concept of exchange rate. Krueger (1983) describes the exchange rate as the price for which a unit of the national currency is exchanged for foreign currency. Exchange rate regimes range from fixed to freely floating exchange rates. The exchange rate policy a country adopts should depend on the current economic situation, size of the economy, the types of exchange rates other countries are using, and the long term economic policy goal.

Under a fixed exchange rate, the local currency is either pegged to another currency or a basket of other currencies. The main goal of this system is to achieve stability in the value of currency through fixing it against a stronger and more stable currency (or currencies). The main advantage of this system is that the exchange rate does not fluctuate due to market conditions, and therefore creates a stable and predictable business climate for investments and international trade. However, the main drawback of a pegged exchange rate regime is that it is difficult for governments to conduct independent monetary policy and to liberalize capital markets at the same time (Thirlwall, 2003).

In contrast, a flexible exchange rate regime is allowing the value of a particular currency in the exchange market to fluctuate based on the supply and demand for that currency. Ghosh, Gulde, & Wolf (2002) indicate that, one of the advantages of the floating regime the regime is the automatic adjustment of balance of payments, since a deficit or surplus is corrected by appreciation or depreciation of the currency. The main disadvantage of this system is the exchange rate volatility. The exchange rate can appreciate substantially and therefore be unmanageable for the

(10)

9 tradable goods sector. On the other hand, when the currency depreciates, it can lead to extreme inflation by raising the domestic prices of import products.

1.1.2 The Real Effective Exchange Rate

Section 1.1.1 provided a brief analysis of the concept of the exchange rate. This section will provide a definition and a brief analysis of the real effective exchange rate, since the latter will be applied in the analysis for Suriname. In order to derive the real effective exchange rate, we first need to take a look at the real exchange rate. M. Chinn (2005) presents a model in which the

real exchange rate plays a central role. The most standard definition of the real exchange rate is 𝑞_𝑡 ≡ 𝑠_𝑡− 𝑝_𝑡+ 𝑝 ∗_𝑡 (1),

where q is the real exchange rate. Furthermore, s is the log nominal exchange rate defined in units of home currency per unit of foreign currency. In addition, p and p* are the domestic and foreign price, respectively. Also, M. Chinn (2005) assumes the price index p is a geometric average of the prices of traded (𝑝_𝑡𝑇_{) and non-traded (𝑝}

𝑡

𝑁_{) goods. This provides for the price index in the home}

country.

𝑝_𝑡= 𝛼𝑝_𝑡𝑁_{+ (1 − 𝛼)𝑝}

𝑡𝑇 and 𝑝𝑡∗ = 𝛼∗𝑝𝑡𝑁∗+ (1 − 𝛼∗)𝑝𝑡𝑇∗ (2)

for the foreign country, where lowercase letters denote logged values. So, substituting (2) into (1) and adjusting yields: 𝑞_𝑡 ≡ (𝑠_𝑡− 𝑝_𝑡𝑇_{+ 𝑝}

𝑡𝑇∗) − 𝛼(𝑝𝑡𝑁− 𝑝𝑡𝑇) + 𝛼∗(𝑝𝑡𝑁∗− 𝑝𝑡𝑇∗) (3)

Equation (3) specifies that the real exchange rate can be expressed as the sum of three components; the relative price of tradables at home versus abroad, the relative price of non-tradables in terms of tradables in the home country, and the matching relative price in the foreign country. Moreover, M. Chinn (2005) indicates that in case the weights of non-tradables in the aggregate price index at home and abroad are identical, the second and third terms can be collapsed into a relative price of non-tradables and tradables between countries, viz: 𝑞𝑡 ≡ ( 𝑠𝑡− 𝑝𝑡+ 𝑝 ∗𝑡) − 𝛼(𝑝̂𝑡𝑁− 𝑝̂𝑡𝑇) (4)

where, the circumflex denotes the intercountry log difference. Hence, if one assumes the law of one price holds for all goods, and consumption basket are the same, then −𝑝𝑡, +𝑝 ∗𝑡, 𝑝̂𝑡𝑁 and 𝑝̂𝑡𝑇

are zero, and PPP holds. So, the real exchange rate is constant. When calculating the real exchange rate, one must also decide which price indices to use. M. Chinn (2005) notes one is often faced with a set of trade-offs, in deciding which price measure is the most applicable in order to calculate the real exchange rate. In reality, one only has a limited choice of price deflators such as, consumer price index (CPI), producer price index (PPI) and whole sale price index (WPI). M. Chinn (2005)

(11)

10 indicates that the CPI is thought of as weighting fairly heavily non-traded goods, for example consumer services. However, PPI and WPI exclude non-trade retail sales services. Furthermore, M. Chinn (2005) mentions that in a two country world, the calculation of the real exchange rate using CPI’s will in theory yield the correct estimate of inter-country relative prices. Although, M. Chinn (2005) stresses that it is unclear of what interest the ratio of the relative non-tradable price between countries is, as it solely relates to the condition of internal balance in the two countries. If the goal is to measure the relative price of goods and services that are tradable, M. Chinn (2005) prefers PPI’s or WPI’s as deflators for the real exchange rate. The main limitation of using these indices is that there is noticeably more difference in how these price series are established across countries, than for the comparable CPI. As M. Chinn (2005) note, this is of specific distress because difference in the real exchange rate can happen even if the law of one price holds for individual goods, if the weights on those two goods diverge in the respective consumption basket. Another likely disadvantage M. Chinn (2005) mentions of using existing indices of tradable goods is that PPI’s or WPI’s may incorporate a large part of imported intermediate goods, such that the resulting real exchange rate is not a good measure of competiveness. In this thesis therefore the CPI is used to calculate the RER for Suriname with its major trade partners.

Since the real exchange rate only relates to two countries it does not deal with the fact that countries trade with multiple partners. The measure used that does take this into account is the real effective exchange rate (REER), which includes real exchange rates of one country with several other countries. By far the most usual means of calculating the real effective exchange rate is to weight the currencies by trade weights. Contemplate a geometrically weighted average of bilateral real exchange rates as, 𝑞𝑡𝑅𝐸𝐸𝑅 ≡ ∑ 𝑤𝑗𝑞𝑡

𝑗 𝑛

𝑗=1 (5),

where 𝑞_𝑡𝑗 denotes the log real exchange rate relative to country j. The easiest way to compute 𝑤_𝑗 is to use weights build upon the sum of exports and imports, indicated as a share of total exports and imports (bilateral relative total trade volumes). However, one problem M. Chinn (2005) notes with this perspective is that there is no theoretically suggested way in which to explain for changing trade flows. At last, once one has resolved on the theoretically compelling weighting scheme M. Chinn (2005) mentions, one resists the situation of enable for time variation. Also, in a way M. Chinn (2005) notes that trade flows adjust over time. Thus, in general the matching trade weights should evolve. The more instant the change of trade structures, M. Chinn (2005) states the more plausible it is that a permanent weight index will misinterpret the effect of exchange rate

(12)

11 fluctuations. Moreover, M. Chinn (2005) mentions that, one can either permit the weights to modify over time deliberately or discretely and occasionally, with the option relying in large share on the trade-off between appliance and accuracy. In conclusion, given the slow movement of trade flows, M. Chinn (2005) mentions chained base year weighted indices as probably a good approximation to the ideal indices. In this thesis the weights will be modified over time every five years.

1.1.3 Drivers of the Real Effective Exchange Rate

Because of their robust effect on the current account and some macroeconomic variables, real effective exchange rates are among the most important price competiveness measures in an open economy. B. Algieri (2013) remarks, that it is crucial to comprehend the main drivers of exchange rates whose changes impact the interactions of households, firms and financial institutions. As B. Algieri (2013) notes, a correct examination of real effective exchange rates is particularly important for natural resource driven economies and transition countries because the vulnerability to supply and adjustment shocks, which affect the particular exchange rate, has a big influence on the real economy. The research on the evaluation of all drivers of the real effective exchange rate is too sizeable for a comprehensive review. However B. Algieri (2013) indicates that the empirical literature on the determinants of the real effective exchange rate can be categorized in several strands.

With respect to the first strand, B. Algieri (2013) explains the dynamics of the real effective exchange rate (REER) in the framework of the Balassa-Samuelson effect, where productivity changes play a central role. If a certain country experiences growth in the productivity of the tradable sector relative to its trade partners, the countries particular REER would probably appreciate. The idea is that for given prices of tradables (small country assumption), such productivity extension would bring an increase in real wages. Hence, if wages are similar across sectors, prices would rise in the non-tradable sector, hence affecting the REER.

The second strand relates to the interest rate, B. Algieri (2013) states that the asset market approach links REER dynamics to shifts in interest rate spread. For instance, a growing interest differential with other nations triggers a real effective exchange rate appreciation, since favorable interest rates in an economy offer investors a higher return relative to other nations. Therefore, B. Algieri (2013) implies that this attracts foreign capital and causes the REER to rise. As a remark, B. Algieri (2013) notes that for developing and transition economies interest rate differentials are

(13)

12 subject to measurement issues.

A third strand of literature that B. Algieri (2013) mentions, emphasizes the importance of fiscal and monetary policy changes. If there is an expansive fiscal policy, interest rates, leaving monetary policy unchanged, will rise and the REER will increase. More clearly, increased government spending may increase spending in goods of the non-tradable sector (𝑝̂_𝑡𝑁_{in equation}

4), enhancing their prices and putting increasing (downward) pressure on REER. Also B. Algieri (2013) notes that, when a fiscal deficit grows, it can be conducted both by an increase in interest rate and by a fall in financial credibility. Put differently, increasing government consumption can reduce reliance in the continuality of public and monetary policies inciting a fall in capital influx and a real depreciation. From a monetary policy perspective, B. Algieri (2013) indicates that domestic Central Banks can regulate the REER through foreign reserves. In order to restrain the appreciation of a currency a Central Bank piles up foreign reserves and to avoid a depreciation it reduces foreign reserves. This operation is more successful when foreign exchange involvement is not sterilized.

A fourth strand of the literature links trade openness and REER movements. In this situation B. Algieri (2013) points out that trade openness can be estimated as a measure for private sector policy. A more depreciated REER is likely to be linked with a more open trade regime. B. Algieri (2013) explains that the trade openness variable measures trade restrictions and catches how such strategies impact the REER through their influence on the price of non-tradable goods. For example, a raise in import tariffs increases the price of imported goods, which can alter prices of non-tradable goods through income and substitution effects. The adverse income effect from higher import prices may lessen demand for all goods and services in the economy, setting downward pressure on the prices of goods produced in the non-tradable sector. The downward pressure can result in a depreciation of the REER. The substitution effect would improve the price of non-tradable goods and could lead to the appreciation of the REER. B. Algieri (2013) states that, some papers claimed that the substitution effect is probably to dominate; hence, a strengthening of trade barriers would lead to an appreciation of the real effective exchange rate, and a more open trade regime would lead to a depreciation of the REER.

In the empirical literature on the impact of mineral prices on the real effective exchange rate, as described in section 1.2, many of the above mentioned drivers of the real effective

(14)

13 exchange rate are used as control variables, next to mineral prices. In this thesis however, as indicated the simple model by V. Bodart et al (2012) without control variables will be used.

1.1.4 The Relationship between Mineral Prices and the Real Effective Exchange

Rate

Next to the above mentioned variables, also mineral prices may have an impact on the real effective exchange rate. Mineral exports have always fueled the economies of Latin America and the Caribbean, supplied its public budget, and assisted as the leading link to world markets. E. Sinnott et al. (2010) indicate, that in countries where natural resources (minerals) can be produced at a lower cost, relative to the marginal cost of production elsewhere they generate large economic rents (profits) for the specific country, in particular when the mineral prices increase. E. Sinnott et al. (2010) point out two major consequences of an increase in mineral prices on the relative incentive structure in the economy. First, it improves the return to production of the resource relative to other goods in the tradable sector. Second, to the extent the resources are exported, the REER appreciates. E. Sinnott et al. (2010) states that, both of these effects lessen the inducement to invest in producing other goods in the tradable sector, resulting in a skewed production and export structure, generally referred to as Dutch disease. The latter attributes the changes of the REER through mineral price shocks. What is more, the specific real effective exchange rate shows a trend to appreciate as a consequence of either one or a combination of the following occurrences; (i) an upsurge in domestic absorption1_{and permanent income; (ii) an increase in the price of}

non-tradable goods; (iii) an increase in relative prices in the non-tradable sector and (iv) a boost in foreign capital inflows, indicated by B. Algieri (2013).

1.2 Empirical Literature Review

1.2.1 Introduction

The empirical literature on the impact of mineral prices on the real effective exchange rate is of very limited size. Table 1 summarizes three different studies done by various authors about the impact of mineral prices on the real effective exchange rate of different countries. Each study will be elaborated in turn in the next three subsections.

1_{Absorption is generally defined as; the sum of households consumption, gross investment, and government}

(15)

14

Table 1: Overview of the Empirical Studies

Countries Determinants Period Test

Peru

M. Tashu (2015)

Price proxy of commodities Relative productivity

Primary public consumption Foreign net liability

Quarterly data 1992-2013 Co-integration DOLS FMOLS 2SLS South Africa S. Ngandu (2015) Price of gold

Non-gold terms of trade Capital flow Forex reserves Gov. expenditure Technical productivity Tariff Quarterly data 1970-1995 Co-integration Panel of developing Countries V. Bodart et al. (2012)

Price of leading commodity Monthly data 1988-2009

Panel Co-integration DOLS

FMOLS

Source: Author’s research

1.2.2 Peru’s case

M. Tashu (2015) investigates the hypothesis of ‘’commodity currency’’ for the Peruvian sol. More specially, it investigates the drivers of Peru’s real effective exchange rate. Also, M. Tashu (2015) notes that, for commodity dependent economies like Peru, the essential step in estimating the equilibrium real effective exchange rate is to aim to see if the Peruvian real effective exchange rate is primary determined by the real prices of key export commodities, hence the term ‘’commodity currency’’. The results suggest that the real price index of minerals does not explain the movement of the real effective exchange rate of Peru, so the sol is not a commodity currency. In addition, M. Tashu (2015) indicates that possible factors that could have weakened the relationship between the mineral prices and the real effective exchange rate may include large profit repatriation by foreign mining companies and active foreign exchange intervention. The analysis of M. Tashu (2015) shows that, the REER is strongly correlated with relative productivity

(16)

15 and primary public consumption. However, the real effective exchange rate is only partially correlated with the stock of foreign net liability, which shifts from positive to negative in 2007.

1.2.3 South Africa’s case

S. Ngandu (2005), evaluates various papers which discuss the link between mineral prices and the real effective exchange rate of South Africa, the researcher notes that any changes in the variables that affect a country’s internal and external equilibrium, will lead to fluctuations in the equilibrium real effective exchange rate. So, any improvement in these variables will lead to an appreciation of the REER. The terms of trade is disaggregated into the major export mining commodity gold and non-gold terms of trade to capture the relative importance of substitution and income dominance. The trade policy variable is a proxy of custom receipts plus surcharges, divided by imports. Other variables are; long-run capital flow, gross reserves of the central bank and, total government expenditure, where theoretically government expenses in the non-tradable sector alone lead to an appreciation of the REER. The comparative measure is the technical productivity levels between South Africa and her trade partners. The co-integration method and single equation error correction models are used to analyze both the short-run and long-run determinants. All these variables including mineral prices where positive and significant determinants of the REER of South Africa.

1.2.4 Panel of Developing Countries’ case

V. Bodart et al (2012) state that, whether natural resources in developing countries is a curse or a luck is still an open debate. In their paper, V. Bodart et al (2012) provide new empirical estimates of the effects of commodity price fluctuations on the real effective exchange rate for a panel of only developing countries. However, V. Bodart et al (2012) formulate a simple relationship between the real effective exchange rate and the real price of the leading export mineral commodity of the country. The use of one single explanatory variable is motivated by the fact that other traditional variables such as, real interest rate differentials or net foreign asset accumulation are considered to be irrelevant for small developing countries. The result of this study is that also here the real effective exchange rate appreciates when the price of the leading export commodity increases. The appreciation is even stronger the larger the share of the specific commodity.

(17)

16 In sum, just like M. Tashu (2015), S. Ngandu (2005) and, V. Bodart et al (2012) the co-integration methodology will be used to analyze the impact of mineral prices and real effective exchange rate of the selected panel countries. In particular, the simple model used by V. Bodart et al. (2012) will be applied in this thesis to the case of Suriname.

(18)

17

2. Trends of Selected Variables

This chapter begins with a brief analysis of the exchange rate development of Suriname. This is followed by, a brief analysis of the real effective exchange rate of Chile, Guyana, Mauritania, Peru and, Suriname. Thereafter, the movement of the prices of Alumina-oxide, gold copper, iron and, zinc are explained.

2.1 The Nominal Exchange Rate Developments

In the period 1940-1994, the Surinamese guilder (Surinamese dollar was introduced in 2004) was fixed to the US dollar, while being free floating to other currencies (CBvS, 1995). The fixed rate was until October 1992 Sf 1.80 in exchange for US $1. Because of the economic crisis in the early 1990s, the government of Suriname decided to implement a structural adjustment program to stabilize fiscal and monetary imbalances. The Central Bank of Suriname introduced a multiple exchange rate system in October 1992. Different rates were used as official exchange rates, namely the auction, the bauxite, the fuel, the rice, and the banana exchange rate. In addition, there was a black market exchange rate. During this period, the national currency depreciated significantly. The Central Bank ended the multiple exchange rate system in 1994 (CBvS, 1996). By then, US $1 was traded for Surinamese guilder 150.75. In 1995 the central bank made an effort to stabilize the exchange rate through intervention. The official exchange rate remained stable throughout the years 1997 and 1998, while in 1999 another devaluation was inevitable. On 1 January 1999, the Central Bank of Suriname traded US $1 for 710 Surinamese guilders.

In spite of the fact that further adjustments (US $ was officially traded for Sf 750,-) were meant to narrow the gap between the official and the black-market exchange rate, the public had no confidence in the macro-economic policies. There was a change in administration in 2000. By that time the exchange rate was Sf 1,173 per US $. In the same year, this rate was adjusted with 88% to Sf 2.200 per US $. In this period there was no stabilization in the currency market (CBvS, 2000). In 2000 the Central Bank introduced an upper bound of Sf 2,400 and a lower bound of Sf 2,000. After several adjustments to the upper and lower bound, the Central Bank of Suriname finally got the exchange rate under control in 2001. However, because of wage increases for the public servants, the exchange rate got out of balance. In the last quarter of 2002 the exchange rate was stabilized at Sf 2,800 per US $ (CBvS, 2003).

(19)

18 Surinamese community had a lot of confidence (CBvS, 2004). From 2004-2010 the exchange rate remained stable at SRD 2.78 per US $ as shown in figure 1. However, when a new government was elected in 2010, the SRD devalued by 20% to SRD 3.35 per US $. The purchase and sales volumes of US dollars on the local foreign exchange market declined during 2014. This also applied to the purchase and sales volumes of euro's which was partly explained by lower demand due to lower current transfers from the Netherlands (CBvS, 2014). In November 2015, the Surinamese dollar was devalued again to SRD 4.04 per US $, according to the Central Bank the reason was that Suriname had experienced a severe commodity price shock in recent years, the gold price fell by more than 55% from its peak, while the export price of oil fell by almost 60% in 2014. As a result, the foreign currency inflow from Suriname and the Central Bank reserves dropped significantly. In the year 2016, the Central Bank of Suriname tried through currency auctions to get the exchange rate under control. Currently, the official rate is SRD 7.41 per US $.

Figure 1: Suriname, with Exchange Rates Trading Partners

Source: Author’s calculation based on the exchange rate with the US$ Converted exchange rates

Figure 1 depicts several nominal exchange rates with the selected trading partners of Suriname. As the sections above describes, the Surinamese currency was de facto fixed to the US$. The currencies from Canada, China, Netherlands and Trinidad & Tobago were all converted to the

(20)

19 US$ to obtain exchange rates of these currencies with respect to the Surinamese currency. As shown in figure 1, the exchange rate of Suriname between, both the Chinese Yuan and the Trinidadian dollar were stable, while going upward from 1999 to 2011. The three other exchange rates depict several fluctuations. Between 1998 and 2010 the currency of Suriname depreciated considerably with respect to the US$, the CAN$ and the T&T$.

2.2 Developments of the REER

Figure 2: The Real Effective Exchange Rate of Suriname

Source: Author’s calculation

The real effective exchange rate is the nominal effective exchange rate (NEER2) adjusted for inflation. This real effective exchange rate is calculated for Suriname. The major trading partners that are included in the currency basket are: Canada, China, Netherlands, Trinidad & Tobago and USA. In the methodology section of this thesis the calculation of the REER for Suriname is further discussed. An overview of the development of the real effective exchange rate of Suriname is depicted in figure 2. An upward movement of the REER index represents an appreciation of the Surinamese currency in real terms relative to the currencies of its major trading partners. A downward movement indicates a depreciation of the Surinamese currency in real terms

(21)

20 with respect to the currencies of its major trading partners. After a period of real appreciation of the Surinamese currency during 1980-1993, a steep real depreciation is evident between 1993 and 1994. This period coincides with the implementation of a structural adjustment program by the government. The reform of the foreign exchange regime has been an integral part of this program. The monetary authorities practiced multiple exchange rates during this period as a transitory phase to a managed floating exchange rate regime. Since then, the Surinamese currency followed a upward trend, implying a real appreciation against the currencies of its major trading partners, on average. An appreciation of the Surinamese currency in real terms implies a loss of competitiveness, which is damaging for the export sector in the long run.

Figure 3: The Real Effective Exchange Rates of Panel Countries

Source: http://bruegel.org/publications/datasets/real-effective-exchange-rates-for-178-countries-a-new-database/

In addition, figure 3 shows a CPI based real effective exchange rates for the other panel countries for the period 1980-2011 normalized to 2005=100. An upward movement of the REER index represents an appreciation of the panel countries’ currency in real terms relative to the currencies of their major trading partners. A downward movement indicates a depreciation of currency of the selected countries in real terms with respect to the currencies of their major trading

(22)

21 partners. Some countries like, Chile and Guyana experienced a steep appreciation in the early 1980s. After that, Chile, Guyana and, Mauritania encountered depreciation between the early 1980s and 1990s. On the other hand, Peru experienced an appreciation in the late 1980s. After the mid-1990s all the four selected countries encountered more or less a constant real effective exchange rate.

2.3 Developments of Mineral Prices

This section begins to define the Manufactures Unit Value Index. In addition, section 2.3 presents the mineral prices of the main export commodities for the four panel countries that are used next to Suriname. After that, the price development of Aluminum-Oxide is shown for Suriname.

V. Bodart et al. (2013) state that in order to show correctly the relationship between real exchange rates and mineral prices, mineral prices have to be denoted in real terms. The use of the Manufactures Unit Value Index (MUV) as a deflator is common in the commodity price literature. MUV is the unit value index denominated in US$ of manufacturing exports from 20 developed nations with country weights based on the nations’ total 2005 exports of manufactures. The base year is 2005. This MUV index deflator for mineral prices is provided by the World Bank database.

Figure 4 indicates, the annual selected commodity prices series, normalized (2005=100) and deflated by the manufacture unit value index. This manufacture unit value index is shown in appendix 1. Gold, Copper, Iron and, Zinc are the selected main export commodities for Chile, Guyana, Mauritania and, Peru. In addition, M. Stuermer (2018) specifies that large moves in mineral prices happen at small frequency. Moreover, price changes have essentially been driven by commodity demand shocks. It demonstrates that shocks to global demand due to, for example, phases of prompt industrialization or economic crisis have always steering substantial booms and busts in mineral prices (M. Stuermer, 2018).

(23)

22

Figure 4: Mineral Prices (2005=100)

Source: http://www.imf.org/external/np/res/commod/index.aspx

Figure 5 denotes, the annual Alumina-Oxide price deflated by the MUV index with 2005 as base year. This manufacture unit value index is shown in appendix 1. Over time the world market price of Alumina-Oxide experienced several booms and busts. Taking the first oil crisis in 1973 into consideration, A. Wzorek et al. (2017) note that the price of many commodities rose significantly, including the price of Aluminum-Oxide in the 1980s. Furthermore, A. Wzorek et al. (2017) mention that much of the rise in prices was also due to prompt rise in electricity prices. At once, the demand for aluminum has expanded steadily, predominantly in the dynamically flourishing automotive industry. However, A. Wzorek et al. (2017) indicate the second oil crisis, which resulted in the economic crisis coincided with inflation and recession, and the collapse in prices of commodities, including Alumina-Oxide. An additional bust in the global aluminum market was created by the fall of the USSR in 1991. A. Wzorek et al. (2017) indicate, that there was a keen rise in the supply of raw material, which translated into a collapse in the price of Alumina-Oxide. A. Wzorek et al. (2017) mention, that the first boom, often referred to as the ‘raw material boom’, occurred in 2006, when food and raw materials prices increased by more than

(24)

23 100%. However, A. Wzorek et al. (2017) note that at the end of the 2007 crisis, which developed in the American real estate market, this crisis also shifted to the raw material market, which resulted in a steep fall in prices, including Alumina-Oxide prices. The lowest price was recorded in 2009, followed by the price rebound up to 2011 (A. Wzorek et al., 2009).

Figure 5: Alumina-oxide (2005=100)

Source: http://www.imf.org/external/np/res/commod/index.aspx

The above mentioned mineral prices will be used in the regression analysis investigating the impact of mineral prices on the REER for our panel countries.

(25)

24

3. Methodology

To analyze whether mineral prices have an impact on Suriname’s REER, a quantitative panel data analysis, based on the method of V. Bodart et al. (2013) is performed. Data on annual basis from 1980 until 2011 is used, due to restricted data availability for other years. The panel consists of the following countries; Chile, Guyana, Mauritania, Peru and, Suriname. These countries experienced over the period 1980-2011 on average, high mineral rents in percentage of GDP. Table 2 shows, the final set of countries-commodities pair including the average mineral rents as % of GDP over the period 1980-2011.

Table 2: Final Set of Country-Commodity Pairs

Source: https://data.worldbank.org/indicator/NY.GDP.MINR.RT.ZS?end=2011&start=1980

The hypothesis tested in this research is: ‘A positive change in mineral prices leads to an appreciation of Suriname’s real effective exchange rate’. To test this hypothesis, first the REER needs to be obtained for all panel countries.

3.1 Calculation of the REER

As far as can be ascertained, data on the REER for Suriname is not available from local economic institutes. Therefore, this study first calculates this exchange rate relative to Suriname’s major trading partners in the index. The most commonly used price deflator for this purpose is the consumer price index CPI, hence the CPI is used in this thesis.

The REER for Suriname is based on the flow of trade between Suriname and its five major trading partners over the period 1980-2011. Canada, China, Netherlands, Trinidad & Tobago and USA are the main trading partners and account for an important share on the trade balance of Suriname. The exchange rates of the Surinamese dollar with Netherlands guilder or euro, Trinidad & Tobago’s dollar and, and US dollar are data from the Central Bank of Suriname. The exchange rates of the SRD with Canada’s dollar and China’s Renminbi are converted from the respective

Commodity Countries Mineral rents % of GDP

1 Iron Mauritania 14.455

2 Copper Chile 9.241

3 Alumina-Oxide Suriname 8.480

4 Gold Guyana 8.286

(26)

25 exchange rates relative to the US$. The exchange rates are detailed in appendix 2. As mentioned above, the corresponding country’s CPI is utilized as deflator for each respective country. The weights are derived from the share of total trade, thus exports plus import.

The formula used by to compute the REER, is as follows: RER country j = ( Nominal Exchange Rate country j X CPI country j

CPI sur )

REER Suriname = Sum (RER country j X TRADE WEIGHT country j,…), where country j is; Canada, China, Netherlands, Trinidad & Tobago and, USA. For the other panel countries, 2005 based CPI real effective exchange rates from Bruegel is used.

3.2 Estimation Technique

V. Bodart et al. (2011) point out that,due to possible non-stationarity of the series, the most common method for evaluating the dependence of the REER of highly specialized countries on the global price of their dominant mineral commodity is related to co-integration methods. Furthermore, if the real effective exchange rate and mineral prices are co-integrated, that this means that the variables are tied together. Although time-series co-integration methods could be used to test the link between real effective exchange rate and real mineral prices, this is considered not appropriate for this thesis because the number of observations per country would be of a too small size to guarantee that the unit root test reach good power. Hence, panel co-integration methods are used to perform the econometric analysis.

Furthermore, V. Bodart et al. (2013) describe the techniques combining panel and non-stationary series to give rise to three type of methods and tests: panel unit root tests, panel co-integration tests and co-co-integration estimation and inference. In addition, one can determine two generations of panel methods. V. Bodart et al. (2013) explain the first generation as methods that are based on the assumption that panel units are cross-sectional independent. However, it is likely that such an assumption is unrealistic for the REER and mineral prices. Therefore, V. Bodart et al. (2013) mention that the second generation panel techniques, which allow for cross-sectional dependence, is a better estimator. The idea of the second generation panel approximation is to approach the limiting distribution of the tests by calculating tests with smaller blocks of consecutively observed time series and to calculate the empirical distribution. Thus, this procedure does not demand estimation of nuisance parameter.

(27)

26 Panel Unit Roots Tests

To test the model, the first step is the unit root tests to test for the non-stationarity of real effective exchange rates and for mineral prices, just like V. Bodart et al. (2013) do in their paper. Moreover, V. Bodart et al. (2013) note, panel unit root tests proposed by Levin, Lin and Chu (2002) and Im, Pesaran and Shin (2003) are the most common tests in empirical research. The former (LLC) pool the panel series, correct and standardize the t-statistics to make it normally distributed. While, the latter (IPS) do not pool the data but estimate separate augmented Dickey-Fuller unit root tests for cross-section units and average the t-tests. After standardization, this average follows a normal distribution. Although these tests are generally utilized, it has been exposed that they are not consistent in the appearance of cross-sectional dependence, as well as when the cross-sectional dimension is small with respect to the time dimension.

Panel Co-integration Tests

V. Bodart et al. (2013) remark, if the panel unit root tests do not reject the hypothesis that real exchange rates and mineral prices are non-stationary, the following step is to check whether the series are integrated. Also, V. Bodart et al. (2013) indicate, that if the variables are co-integrated, that this means that the real effective exchange rates of the countries are tied to the mineral prices. In a panel setting, V. Bodart et al. (2003) mention that, the usual tests are those that belong to the residuals based co-integration tests family. Cross section independence is the assumption.

Panel Co-integration Estimate and Inference

In addition, V. Bodart et al. (2013) mention, that if the REER and mineral prices are found to be non-stationary, the next step is to estimate the co-integration coefficient 𝛽 in the main model. Also, V. Bodart et al. (2013) note that the use of normal Ordinary Least Squares techniques leads to counterfeit regression when the series are non-stationary and, therefore, particular panel co-integration techniques have to be used. V. Bodart et al. (2013) specify that in the case of homogeneous and near-homogeneous panels, the coefficient of co-integration can be estimated by a fully modified estimator. In particular, V. Bodart et al. (2013) state that this method challenges the plausible problem of endogeneity of the regressors as well as the autocorrelation of residuals. Also, V. Bodart et al. (2013) propose a dynamic least square estimator (DOLS), alternatively. This

(28)

27 approximation method has the advantage of computing convenience and is parametric. Section 4, presents the outcomes secured with both DOLS and FMOLS estimators.

3.3 Data and Model

The variables used in this study were selected based on empirical studies in particular V. Bodar et al. (2013) and the nature of the Surinamese economy. The data of the utilized variables are collected from the database of the World Bank, Bruegel, IMF and, Central Bank of Suriname. Data on an annual basis from 1980-2011 was utilized.

The main model used in this research:

REER𝑖,𝑡 = 𝛽MINERAL_PRICES𝑖,𝑡+ 𝑒𝑖,𝑡

MINERAL_PRICES𝑖,𝑡 = MINERAL_PRICES𝑖,𝑡−1+ ℇ𝑖,𝑡

𝑒_𝑖,𝑡 = 𝜆_𝑖′F_𝑡+ U_𝑖,𝑡

F𝑡= F𝑡−1+ 𝜂𝑖,𝑡

REER_𝑖,𝑡= The Real Effective Exchange Rate with 2005 base year, hence a decrease in the REER index is considered a depreciation and an increase in the index is considered appreciation. MINERAL_PRICES_𝑖,𝑡 = Prices of the selected commodities deflated by manufactures unit value

index.

V. Bodart et al. (2013) indicate, that F_𝑡 are unobserved factors and 𝜆′_𝑖 the factor loadings. Furthermore, V. Bodart et al. (2013) note, that this method imposes a factor structure on 𝑒_𝑖,𝑡 to catch the cross-sectional dependence.

3.4 Expectations

Based on literature, the expectations are that by using the methodology described above a possible quantitative conformation will be found that mineral prices has a positive impact in the long run on the real effective exchange rate. Given that an increase in the REER index is equivalent to a real appreciation of the domestic currency or a real depreciation of the foreign currency and a decrease in the REER index is considered a real depreciation of the local currency or a real appreciation of the foreign currency.

(29)

28

4. Results

4.1 Data Analysis

In this part, the study employs regression models with the software Stata, to test the effects of mineral commodity prices on the real effective exchange rate of selected countries, including Suriname. As mentioned in section three, the variable mineral price is tested as a determinant of the real effective exchange rate in subsequent regressions. The Real Effective Exchange Rate is measured as an index with 2005 as base year.

Stationarity analysis

In order to test for stationarity in this thesis, two methods will be applied, based on Bodart et al. (2013). The two tests are the Levin-Lin Chu (LLC) unit root test and the Im-Pesaran-Shin (IPS) unit root test. Both methods will be applied to test for a unit root, first in the data for mineral prices and then in the data for the real effective exchange rates. The descriptive statistics of the real effective exchange rate and mineral price variable are presented in Appendix 3. The results of the stationary tests of mineral prices is revealed in Table 3 & 4. The output of Stata show, that each of the two tests examined (LLC & IPS) do not reject the null hypothesis in the panel. For the LLC tests, this outcome remains valid at a significant level of 10%. Also, in the IPS test, the null hypothesis of unit roots is rejected at significant levels of respectively 1%, 5% and, 10%. Consequently, there is sufficient confirmation to consider mineral prices as non-stationary. Moreover, V. Bodart et al (2013) indicate that this result that they also find is compelling as it differs from the standard suggestion that mineral prices are somewhat foreseeable due to seasonality movements.

(30)

29

Table 3: LLC Unit Root Tests for Mineral Prices

Source: author using Stata3

Table 4: IPS Unit Root Tests for Mineral Prices

Source: author using Stata

The panel unit root tests for the real effective exchange rates are presented in Table 5 and 6. In addition, the IPS unit root test do not reject the null hypothesis. However, the LLC test affirm that the real effective exchange rates obtain a panel unit root. Just like V. Bodart et al. (2013) argue, PPP is therefore rejected.

(31)

30

Table 5: LLC Unit Root Tests for Real Effective Exchange Rates

Source: author using Stata4

Table 6: IPS Unit Root Tests for Real Effective Exchange Rates

Source: author using Stata

In sum, for mineral prices, both tests find that there is no unit root. For the real effective exchange rates, the IPS test finds no unit root, while the LLC test does find a unit root. Next, we test for co-integration between the mineral price series and the real effective exchange rate series.

(32)

31 Co-integration analysis

After the unit root test, next the co-integration test is performed. The Westerlund test for co-integration tests whether mineral prices and the real effective exchange rates are cointegrated in the long-run. The results of Westerlund test for panel co-integration are presented in Table 7. The test reject the null hypothesis of no co-integration at 10% significance. Hence, the existence of a long-run relationship between the panel countries’ real effective exchange rate and selected mineral prices is supported by the data. Appendix 4 shows a residual graph to understand and improve the regression model.

Table 7: Co-integration Analysis

(33)

32 Evaluation of co-integration relationship5

After the outcome of the former subpart, this section now continues to the approximation of the co-integration correlation. The two methods that will be applied are the dynamic ordinary least squares regression and the fully modified ordinary least squares regression. Table 8 & 9 present the results of both tests.

Table 8: Model 1 – DOLS Results

Source: author using Eviews

No matter what approach is applied,it appears that the mineral price coefficient is positive and significant, as presumed. The model of which the results are presented in Table 8 is well specified and the variable in the model explains about 68% of the variance in the dependent variable. The elasticity of output on the real effective exchange rate stands at 0.6527, suggesting a positive relationship between mineral price and real effective exchange rate. Furthermore, the results show in Table 9 is satisfactorily described and the determinant in the model explains about 20% of the variance in the dependent variable. The elasticity of outcome on the real effective exchange rate stands at 0.1294, proposing that mineral price shocks have a powerful long-run effect on the real effective exchange rates of the panel countries, including Suriname. In both models, the result shows that a 10% increase in mineral price, will lead to respectively 6.80 % and

5_{In table 8 and table 9, the real effective exchange rate and mineral price variables are in logarithm}

Dependent Variable: REER

Method: Panel Dynamic Least Squares (DOLS) Sample (adjusted): 1991 2010

Periods included: 20 Cross-sections included: 5

Total panel (balanced) observations: 100 Panel method: Weighted estimation Cointegrating equation deterministics: C

Fixed leads and lags specification (lead=1, lag=10)

Long-run variance weights (Bartlett kernel, Newey-West fixed bandwidth)

Variable Coefficient Std. Error t-Statistic Prob.

MINERAL PRICE 0.652725 0.195221 3.343510 0.0020

R-squared 0.680816 Mean dependent var 4.606999

Adjusted R-squared 0.070610 S.D. dependent var 0.360791

S.E. of regression 0.347820 Sum squared resid 4.113273

(34)

33 1.29% points appreciation in the real effective exchange rate, ceteris paribus. As the mining sector dominates the Suriname economy, this results in a skewed production and export structure, generally referred to as Dutch disease. Appendix 5 is used to determine if the data set is well-modeled by a normal distribution.

Table 9: Model 2 – FMOLS Results

Dependent Variable: REER

Method: Panel Fully Modified Least Squares (FMOLS) Sample (adjusted): 1981 2011

Periods included: 31 Cross-sections included: 5

Total panel (balanced) observations: 155 Panel method: Weighted estimation Cointegrating equation deterministics: C

Long-run covariance estimates (Bartlett kernel, Newey-West fixed bandwidth)

Variable Coefficient Std. Error t-Statistic Prob. MINERAL PRICE 0.129405 0.033485 3.864567 0.0002 R-squared 0.208532 Mean dependent var 4.658858 Adjusted R-squared 0.181973 S.D. dependent var 0.460917 S.E. of regression 0.416875 Sum squared resid 25.89397 Long-run variance 0.156303

(35)

34

Conclusion and Recommendation

This thesis has examined the impact of mineral prices on the real effective exchange rate of Suriname. It has also first calculated the REER for Suriname during 1980-2011. Next, this thesis has used panel data to investigate the relationship between mineral prices on the real effective exchange rate for a set of countries, namely; Chile, Guyana, Mauritania, Peru and, Suriname. The other countries, next to Suriname are added to increase the number of observations and hence the robustness of the analysis. V. Bodart et al. (2013) state, that whether mineral price booms have a positive or negative effect on the production of mineral exporting countries stays unsure up to date. The results of this thesis show that an appreciation of Suriname’s currency and of the currencies of the other countries is due to an improvement of its mineral price, in particular Aluminum-Oxide in the long run. The outcomes consequently recommend that small developing countries totally concentrated on the export of commodities, in particular mineral commodities are endangered to the so called ‘’Dutch disease’’ effects.

With regard to policy making in Suriname, policymakers should take into account the fact that the mining industry (mineral prices) is a major determinant of the REER. Investing the mining profits in other sectors where Suriname has major competitive advantages, to diversify the economy, more specifically the exports sector will be essential. In a time where Suriname is part of a globalized and integrated world it is important to know which sector is productive, so good choices and trade decisions can be made to improve the welfare and wellbeing of its inhabitants. Also, the mining surpluses can act as a financial buffer in periods with a lot of uncertainty and global recessions.

(36)

35

Bibliography

Algieri, B. (2013). Determinants of the real effective exchange rate in the Russian Federation. The Journal of International Trade & Economic Development.

Bodart, V., Candelon, B., & Carpantier, J. F. (2012 ). Real exchanges rates in commodity producing countries: A reappraisal. Journal of International Money and Finance.

Bruegel. (2015). Publications. Opgehaald van Bruegel: http://bruegel.org/

Centrale Bank van Suriname. (1995). Jaarverslag. CBvS.

Chinn, M. D. (2005). A PRIMER ON REAL EFFECTIVE EXCHANGE RATES: DETERMINANTS, OVERVALUATION, TRADE FLOWS AND COMPETITIVE DEVALUATION. Cambridge, MA: NATIONAL BUREAU OF ECONOMIC RESEARCH .

Ghosh, A., Gulde, A.-M., & Wolf, H. (2002). Exchange Rate Regimes: Choices and Consequences. Massachusetts: The MIT Press.

Im, K. S., Pesaran, M. H., & Shin, Y. (2003). Testing for unit roots in heterogeneous panels. Journal of Econometrics.

(37)

36 International Monetary Fund. (2018). Research. Opgehaald van IMF:

http://www.imf.org/external/np/res/commod/index.aspx

Krueger, A. (1983). Exchange-Rate Determination. Cambridge: Cambridge University Press.

Levin, A., Lin, C.-F., & Chu, J. C.-S. (2002). Unit root tests in panel data: Asymptotic and finite-sample properties. Journal of Econometrics.

Ngandu, S. (2005). Mineral Prices and the Exchange Rate: What Does the Literature Say? Employment Growth & Development Initiative Human Sciences Research Council.

Sinnott, E., Nash, J., & De La Torre, A. (2010). Natural Resources in Latin America and the Caribbean. The World Bank.

Stuermer, M. (2018). 150 YEARS OF BOOM AND BUST: WHAT DRIVES MINERAL COMMODITY PRICES? Federal Reserve Bank of Dallas.

Tashu , M. (2015). Drivers of Peru's Equilibrium Real Exchange Rate: Is the Nuevo Sol a Commodity Currency? . IMF Working Paper .

The World Bank Group. (2018). Data Bank. Retrieved from World Bank: https://data.worldbank.org/indicator/NY.GDP.MINR.RT.ZS?end=2011&start=1980

Thirlwall, A.P. (2003). Trade, the Balance of Payments and Exchange Rate Policy in Developing Countries. Massachusetts: Edward Elgar Publishing, Inc.

UNCTAD. (2014). Trade and Development Report. UNITED NATIONS.

Wzorek, A., Ivashchuk, O., & Wzorek, Ł. (2017). Analysis of the factors influencing the price of aluminum on the world market .

(38)

37

Appendices

Appendix 1

(39)

38

Appendix 2

Source: Centrale Bank van Suriname & Author’s calculation

Exchange Rates: Foreign countryto Suriname

YEAR CAN CHI NED T&T USA

1980 0.002 0.001 0.002 0.001 0.002 1981 0.001 0.001 0.002 0.001 0.002 1982 0.001 0.001 0.001 0.001 0.002 1983 0.001 0.001 0.001 0.001 0.002 1984 0.001 0.001 0.001 0.001 0.002 1985 0.001 0.001 0.001 0.001 0.002 1986 0.001 0.001 0.002 0.001 0.002 1987 0.001 0.000 0.002 0.001 0.002 1988 0.001 0.000 0.002 0.001 0.002 1989 0.001 0.000 0.002 0.001 0.002 1990 0.002 0.000 0.002 0.001 0.002 1991 0.002 0.000 0.002 0.000 0.002 1992 0.001 0.000 0.002 0.000 0.002 1993 0.010 0.002 0.014 0.002 0.013 1994 0.148 0.023 0.169 0.035 0.202 1995 0.322 0.053 0.620 0.076 0.442 1996 0.292 0.048 0.531 0.067 0.398 1997 0.286 0.048 0.459 0.064 0.396 1998 0.267 0.048 0.451 0.063 0.396 1999 0.574 0.103 0.947 0.137 0.853 2000 0.884 0.159 1.301 0.210 1.313 2001 1.393 0.261 1.969 0.350 2.157 2002 1.468 0.278 2.339 0.373 2.304 2003 1.840 0.312 3.160 0.414 2.578 2004 2.075 0.326 3.477 0.432 2.699 2005 2.221 0.329 3.457 0.432 2.692 2006 2.385 0.339 3.487 0.433 2.705 2007 2.523 0.356 3.805 0.432 2.710 2008 2.540 0.390 4.072 0.435 2.710 2009 2.371 0.397 3.862 0.432 2.710 2010 2.631 0.400 3.682 0.429 2.710 2011 3.284 0.503 4.657 0.510 3.250

(40)

39

Appendix 3 REER

MIN. PRICE

Mean

118.0391

96.85345

Median

106.1006

80.24365

Maximum

656.0278

487.1794

Minimum

25.43149

39.94020

Std. Dev.

64.41992

61.23323

Skewness

4.026349

3.528499

Kurtosis

32.11754

20.11781

Jarque-Bera

6084.514

2285.471

Probability

0.000000

Sum

18886.26

15496.55

Sum Sq. Dev.

659838.3

596171.9

Observations

160

(41)

40

Appendix 4

Source: author using Eviews -0.8 -0.4 0.0 0.4 0.8 1.2 3 4 5 6 7 C h il e - 8 0 C h il e - 8 6 C h il e - 9 2 C h il e - 9 8 C h il e - 0 4 C h il e - 1 0 G uy a na 84 G uy a na 90 G uy a na 96 G uy a na 02 G uy a na 08 M a u ri ta n ia 8 2 M a u ri ta n ia 8 8 M a u ri ta n ia 9 4 M a u ri ta n ia 0 0 M a u ri ta n ia 0 6 P er u 80 P er u 86 P er u 92 P er u 98 P er u 04 P er u 10 S ur ina m e 84 S ur ina m e 90 S ur ina m e 96 S ur ina m e 02 S ur ina m e 08

(42)

41

The REER and mineral exporters : case study of Suriname

UNIVERSITY OF AMSTERDAM

FACULTY OF BUSINESS & ECONOMICS