In this first chapter, there are some issues that should be considered as the preliminary basis to acknowledge the hedging ability of gold against inflation and exchange rate fluctuations. Firstly, we highlight the idea of the effects of inflation and exchange rate fluctuations, which hopefully will be helpful to trigger first thought on this study. Then, we will descriptively analyze the economic backgrounds of the selected samples (in this case, Malaysia) which mainly related to the hypothesis such as Malaysia's inflation rate, exchange rate against USD, and Malaysia's gold price. Next, we will figure out the problem statements followed by the study objectives, the significance of the study and the study organization.
1.2 Background of the Study
1.2.1 Inflation Rate
Currency buys fewer goods and services when the general price level rises. Inflation also reflects loss in the purchasing power of money which means a loss of real value in the internal medium of exchange and unit of account in the economy. The main measure of price inflation is the inflation rate, the annualized percentage change in a general price index, normally the Consumer Price Index, over time. Inflation's effects on an economy are various and can be simultaneously positive and negative. Negative effects of inflation include a decrease in the real value of money and other monetary items over time, uncertainty over future inflation may discourage investment and savings, and high inflation may lead to shortages of goods if consumers begin hoarding out of concern that prices will increase in the future. Economists generally agree that high rates of inflation and hyperinflation are caused by an excessive growth of the money supply. Views on which factors determine low to moderate rates of inflation are more varied. Low or moderate inflation may be attributed to fluctuations in real demand for goods and services, or changes in available supplies such as during scarcities, as well as to growth in the money supply. However, the consensus view is that a long sustained period of inflation is caused by money supply growing faster than the rate of economic growth.
High or unpredictable inflation rates are regarded as harmful to an overall economy. They add inefficiencies in the market, and make it difficult for companies to budget or plan long-term. Inflation can act as a drag on productivity as companies are forced to shift resources away from products and services in order to focus on profit and losses from currency inflation. Uncertainty about the future purchasing power of money discourages investment and saving. Friedman (1977) argues that high inflation can give rise to political pressure to reduce it. The monetary authority, however, may or may not be reluctant to lower inflation, resulting in future inflation uncertainty. He further contends that uncertainty could cloud economic decisions, reducing economic growth. Ball (1992) formalizes this relationship in a model of asymmetric information between policy makers and the public. Conversely, Cukierman and Meltzer (1986) suggest the possibility that inflation uncertainty could cause higher inflation as the central bank takes advantage of an uncertain environment to produce inflation surprises to stimulate the economy. This relation may further encourage a central bank's inflationary bias, leading to lower long-run economic growth. And inflation can impose hidden tax increases, as inflated earnings push taxpayers into higher income tax rates unless the tax brackets are indexed to inflation. With high inflation, purchasing power is redistributed from those on fixed nominal incomes, such as some pensioners whose pensions are not indexed to the price level, towards those with variable incomes whose earnings may better keep pace with the inflation. This redistribution of purchasing power will also occur between international trading partners. Where fixed exchange rates are imposed, higher inflation in one economy than another will cause the first economy's exports to become more expensive and affect the balance of trade. There can also be negative impacts to trade from an increased instability in currency exchange prices caused by unpredictable inflation.
1.2.2 Exchange Rate Fluctuation
Depreciation lowers the foreign currency price of exports and should increase export quantity. Export revenue in domestic currency, however, may not rise and can fall. Perfectly inelastic foreign import demand would imply no increase in export revenue. If there is high import content in export production, depreciation could result in higher price of exports. With appreciation, exporters might price to market and lower their domestic currency price to maintain market share. Exporters may also actively hedge in option markets to avoid exchange rate effects. Foreign exchange risk refers to the risk faced due to fluctuating exchange rates. For example, a Malaysian trader who exports palm oil to India for future payments in Rupees is faced with the risk of Rupees depreciating against the Ringgit when the payment is made. This is because if Rupee depreciates, a lesser amount of Ringgit will be received when the Rupees are exchanged for Ringgit. Such risks are quite common in international trade and finance. A significant number of international investment, trade and finance dealings are shelved due to the unwillingness of parties concerned to bear foreign exchange risk. Hence it is imperative for businesses to manage this foreign exchange risk so that they may concentrate on what they are good at and eliminate or minimize a risk that is not their trade. Even if there were a positive effect of depreciation on export revenue, associated exchange risk might discourage exporters and mitigate the positive effect. Exchange risk has become an issue since the collapse of fixed exchange rates in the early 1970s but there is no consensus regarding its impact on export revenue. Exchange risk could theoretically lower export revenue due to profit risk as developed by Ethier (1973). De Grauwe (1988) suggests, however, that exporters might increase volume to offset revenue loss. On the other side of the transactions, importers may seek other sources when facing exchange risk. Broll and Eckwert (1999) note the return on an option to export should increase along with risk. Exchange risk could also alter the currency inventory practices of exporters and importers.
The 1997 East Asian currency crisis made apparent how vulnerable currencies can be. The speculative attacks on the Ringgit almost devastated the economy if not for the quick and bold counter actions taken by the Malaysian government, particularly in checking the offshore Ringgit transactions. It also became apparent the need for firms to manage foreign exchange risk. Many individuals, firms and businesses found themselves helpless in the wake of drastic exchange rate movements. Malaysia being among the most open countries in the world in terms of international trade reflects the degree of Malaysia's exposure to foreign exchange risk.
1.2.3 Gold As a hedge
The asset's returns, which offset the effect of inflation, are termed as hedge against inflation. Different assets play a role as hedging against inflation Bodie (1976). For example society hedge against inflation if it eliminate or reduces the possibility of the real rate of return falling below the specified level or as an asset whose real return is independent of the inflation rate. One of the properties of hedging is to reduce the variability of future wealth Bonnekamp (1978). Hedges is not important for individual only who want to maintain the purchasing power but also important for institutional investors Bodie (1979).
Real investors are concerned about the real values of their assets because their liabilities are linked to inflation. Various physical and financial assets are used as hedging inflation. These are foreign currency, gold, real estate, saving deposits, silver, stock prices, treasury bills and government securities. In the International Conference on Gold Dinar Economy 2007, Tun Dr Mahathir noted that in the case of paper people will have risk in losing their value and also purchasing power. He stressed back that only Gold Dinar really has a value in it. Using monthly gold price data from 1976 until 1999, and cointegration regression techniques, Ghosh et al. (2004) investigate the contradiction between short-run and long-run movements in the gold price and find that the gold price rises over time at the general rate of inflation and hence is an effective hedge against inflation under a set of conditions. When it comes to inflation, the value of gold is considered to be preserved, for its price will increase along with the rise in the general level of prices. In other words, it is believed that changes in the price of gold reflect inflationary pressure. However, the issue is not whether gold hedges against inflation, but how well it does. Each country has its own economic conditions or characteristics. Therefore, this study applies the non-linear model to examine the inflation hedging ability of gold in each country instead of the linear model, in order to find out the more adequacy results.
Having gold as money, or as the basis of the monetary system, meant linking a currency to gold at a fixed price. The behavior of prices was thus taken outside the control of government and central banks, and depended on the gold supply relative to the demand for it. In such a situation an automatic stabilizing mechanism was in place. Suppose that for some reason the price of goods rose relative to gold; this fall in the relative price of gold reduced incentives to produce gold, and also diverted some of the existing stock to non-monetary uses such as jewelries. Conversely, if the price of goods fell there was a rise in the relative price of gold, and thus a stimulus to its production. Hence, a considerable degree of price stability in terms of gold was to be expected. Crucial to this mechanism was gold being the basis of the monetary system. When gold no longer had that role, the automatic stabilizing mechanism, working from changes in the relative gold price, through changes in gold output and use, to changes in the money supply, was no longer in place. That does not mean that gold could no longer be a good store of value or protection against exchange rate change. But whether it is or not depends on different forces. It depends on whether, when currencies weaken, people switch to gold; and on when currencies strengthen; they become more confident about the value of currencies, and switch from gold. Even though gold no longer has any role in the monetary system of any major country, such behavior could still be sensible. For, as de Gaulle pointed out, gold has no nationality and is not controlled by governments. In determining the extent to which gold acts as an exchange rate hedge, it is, therefore, well worth exploring the past, to see how well gold protected against currency fluctuations. That, in itself, is of interest, and it may also be of interest in the future.
1.2.4 Inflation in Malaysia
Figure 1
Low inflation and sustainable GDP growth has been one of the main features of the Malaysian economy in the last two decades. Despite its robust economic growth in 1980s and 1990s, Malaysia's inflation rate had been relatively low by international standards. Even after the severe Asian financial crisis (1997 and 1998) and sharp depreciation of the ringgit in1997-98, Malaysia's inflation rate has been contained at a relatively low level as shown in figure 1.In the early 1970s; Malaysia experienced a single-digit episode of inflation only 2%. During the second half of 1970s, inflation rate gradually increase to 4%. The sharp oil price increase in 1973 and 1974 was the principal reason for the escalation of world inflation in 1973-1974. Consequently, consumer prices in Malaysia began to rise and had reached to double-digit level of 10.56 % by the end of the year of 1973. In 1974, the surge in the oil price by over 230 per cent put strong fuel on inflation, and the inflation rate in Malaysia increased to its record high of 17.32%. A year later, the Malaysian economy slumped into its great recession, with a GDP growth rate of only 0.8% in 1975, compared to8.3% and 11.7% in 1973 and 1974 respectively. On the other hand, inflation rate reduced to the level of 4.5% in 1975.Malaysia experienced a second episode of high prices in 1980 and 1981, which were due mainly to external factors. Oil prices rose by 47% in 1979 and 66% in 1981. As a result, inflation in Malaysia accelerated from 3.6% in 1979 to 6.6% and 9.7% in 1980 and 1981 respectively. Since 1982, inflation rate kept decreasing and amounted to less than 1% in 1985 and 1986. The development of the Malaysian economy was at an important crossroad in 1985. The economic performance of the country had slumped into its greatest recession, with -1.1% and 1.1% growth rate recorded in 1985 and 1986 respectively. From 1990 until 2012, the rate can be said steadily revolve around 1 to 5%.
1.2.5 Exchange Rate Against USD in Malaysia
Figure 2
Malaysian Ringgit (RM) was formerly known as Malaysia Dollar (M$). M$ was created in June 1967 to replace the old Sterling-link Malaysian/Straits Dollar. In year 1971, M$ was linked to Pound Sterling (₤) at fixed rate of 7.4369M$/₤. With floating of Sterling and dismantling of Sterling Area, Malaysia adopted US Dollar with fluctuation range for Effective Rate as intervention currency in place of Sterling in 1972. The intervention of Malaysian Central Bank was to maintain the stability in the value of domestic currency in relation to basket of foreign currencies. Due to devaluation of US Dollar in February 1973, the Official Rate of Malaysian Dollar was realigned to 2.53M$/US$, based on currency's unchanged gold content. In 21 June 1973, Malaysia placed a controlled, floating effective rate In 1975, the Malaysian Dollar was officially changed to Ringgit (RM) and the controlled, floating effective rate was replaced. The external value of Ringgit was determined based on the weighted basket of foreign currencies of the Malaysia major trading partners.
The same exchange rate determination was sustained up till the Asian Financial Crisis 1997/98. During the crisis year, the overvalued Ringgit depreciated sharply against the US dollar by more than 40%. To stabilize the financial market, Malaysia imposed capital control and returned to fixed exchange rate that pegged to US dollar at RM3.80 in September 1998. As part of the economic recovery strategy, Malaysia has committed to export-led growth policy based on maintenance of their undervalued and pegged currencies against the USD. On July 21, 2005, Malaysia responded to China's de-pegging announcement within an hour after the 7-year pegging. Akin to the Chinese policy, BNM allows the ringgit to operate in a managed floating system based on a basket of several major currencies. From 1973 until 1997 as before the major economic crisis started, the exchange rate between RM and USD float steadily around RM2.5 per USD. It rises greatly in 1998 to nearly RM4 per USD and been pegged to RM3.8 from then until 2005.
1.2.5 Gold Price in Malaysia
Figure 3
As clear as can be seen from the figure, the overall pattern of gold price is increasing over time. It rises steadily from 1970, which started at RM110 in 1970, to RM671 in 1979. Then it has a sharp increase in 1980, to RM1333. Dropped a bit to RM877 until 1982, then gold price was steadily revolved around that price. From 2000 until 2011, gold has experienced a great increase in its price, resulted in 353% increment from 2000. This steady increment and sharp for the last decade is what interests us to see whether gold can be used to hedge both inflation and exchange rate fluctuation.
1.3 Problem Statement
In the macroeconomics literature, there are numerous studies and researches conducted to validate this controversial proposition. The problem with inflation and exchange rate fluctuations are something that cannot be avoided. The best thing that any entity could do is to manage them. The problem with this is to choose the best way and the best tool to achieve it. In addition, it was found that most studies are interested on testing the hypothesis in developed countries like European countries, United States and new emerging economy like India and China. However, it is interesting to test the hypothesis in Malaysia because this association is consisted with different level of economic structure countries within the South East Asia region, therefore; indisputably require a concrete analysis to examine the differences.
Figure 4
Even though figure 3 shows stable increment in gold price and somehow offers security for investor, but it is still does not provide enough reason as to why we have to study the ability of gold as a hedging tool against inflation. As can be seen in figure 4, overall, the increment in the price of gold is rather higher than the inflation rate, meaning that the rise in price of gold is higher than the rises in the price index. So, it come to the understanding that rises in price of gold can offset the rise the price of good. This also supported by figure 5 where log of price of gold from 1970 until 2011 totally offset the consumer price index for Malaysia. But this is just the overall idea of the hypothesis. Technically and empirically, it has to be proven.
Figure 5
Figure 6
Same as to hedge exchange rate fluctuation. By seeing figure 3, one could simply conclude that gold can easily hedge exchange rate fluctuation in Malaysia. As shown in figure 6, in overall, changes in gold price is clearly offset the changes in exchange rate. According to figure 6, if gold can really hedge against exchange rate fluctuation, a lot of entities would be benefited by seeing how much difference between both changes. Producers, for example, would be making much profit if the quotation of material were quoted in gold when the RM depreciates and the price of gold is much higher than the depreciation. But, as stated above, this still have to be proven empirically.
1.4 Objective of the Study
The general objective of this study is to show the hedging ability of gold. Likewise, there are two other specific objectives that may help to strengthen this study, namely;
1. To examine the hedging ability of gold against inflation rate in Malaysia,
2. To examine the hedging ability of gold against exchange rate fluctuation in Malaysia.
1.5 Significance of the study
As the objective of the study is to prove the hedging ability of gold against inflation and exchange rate fluctuation in Malaysia, hence, this study definitely will give benefits to both macro and micro level of economics in Malaysia. At macro level, policy makers could warn or suggest to people to hold gold if it is proven its ability to hedge against inflation and/ exchange rate fluctuation. For example, if it is proven that gold can hedge against both, policymaker could launch campaign on suggesting household and firms to hold few gram of gold per unit of economies. The significant for that reason is remarkable as this issue is related to the macroeconomic development through the fiscal policy framework. At micro level, individuals or households could diversified their portfolio in a better and safe way as gold, if proven, could hedge against inflation and exchange rate risk. Firms that deal with international trading which are prone to the exchange rate risk could minimize their risk by holding gold if proven that it can hedge against exchange rate fluctuation. The findings of this study will benefit those who are directly and indirectly involve in the decision making process likely the politicians, economists, policy makers, firms and households. It is crucial to test the objectives of this study especially for the betterment of Malaysia's economic growth and its people well-being.
1.6 Scope of the study
In this study, we will mainly investigate the hedging ability of gold. Interestingly, to meet the objectives, we will use Yuan and Kuang (2011) models and estimate the selected model using the Threshold Vector Auto Regressive test approach to capture the long run and short run effects of the stated objectives. Although there are various techniques to empirically analyze the objectives and estimate the models, however this study only limited to the proposed undertakings.
1.7 Organization of the study
The study is separated into 5 main chapters. Next, the Literature Review will contain some shortcomings of the hypothesis, the frameworks and empirical evidences on the hedging ability of gold against inflation and exchange rate fluctuation. In the third chapter, Methodology and Estimation Procedures, we will disclose the models that will be used in estimating the objectives and state the analytical and diagnostic procedures. After estimate the model, we will show the results and discuss them in the fourth chapter, Results and Discussion. Here, we will display all the analysis that we obtain in proper tables and elaborate further the results and necessarily relate the findings with previous studies. Finally, we will conclude the findings in the last chapter, Summary, Conclusion and Recommendation. We will make general conclusion based on the findings, state few implications on the policy and recommend some improvements for the betterment of future study.
CHAPTER TWO
LITERATURE REVIEW
2.1 Introduction
In this chapter, we will analyze and summarize briefly the findings of the previous studies regarding gold's hedging ability hypothesis and the related issues of this study outright. In order to make the literature comprehensible, the literature will be divided into two main sections.
2.2 Hedging Ability of Gold against Inflation
Firstly, we will discuss on the literatures related to the relationship between inflation and gold price. Various articles published in both academic journals and the financial press show a relationship between the price of gold and actual inflation over time. For instance, in the academic literature, Sherman (1982), Young and Khoury (in Khoury, 1984, pp. 355-358), Haubrich (1998), Jaffe (1989), and McCown and Zimmerman (2006) all find a significant relationship between measures of inflation and gold prices. Ghosh, Levin, Macmillan, and Wright (2004) and Worthington and Pahlavani (2007) present evidence of long run cointegration between gold and inflation. In contrast, Tully and Lucey (2007) using a power GARCH approach, do not find a significant relationship between inflation and gold. The findings of prior studies that prove the effective inflation hedge of gold are almost consistent. However, there is a constraint on research technique employed because most of the literatures utilize the linear model to explore the relationship between the gold return and inflation. They thus ignore the possible non-linear relationship between these two variables. It is known that the gold price and inflation might fluctuate with business cycle, which cause the non-linearity or asymmetry of the model fitted for their relationship. If asymmetric phenomenon is not taken into account in the model estimation or is not verified before the estimation, the empirical findings might be biased.
The idea of gold hedging against inflation is not new, which is virtually found with related papers like "gold is an asset of "safe havens" against the debasement of paper money", "gold is leading indicators of inflation" or "gold is an inflation hedge" and so on. Mahdavi and Zhou (1997) test the performance of gold and commodity prices as leading indicators of inflation with cointegration and vector error correction model (VECM) over 1958-1994. Their findings show that the strength of the gold price signaling inflation may vary depending on the time span being examined. Ranson and Wainright (2005) conclude that the price of gold is the superior predictor of the next year inflation. Laurent (1994), Harmston (1998) and Ghosh et al., (2004) study the relationship between the gold price and wholesale price and find that gold acts effectively as a long-run inflation hedge in U.S., Britain, France, Germany, and Japan. Using monthly gold price data (1976-1999), and cointegration regression techniques, Ghosh et al. (2004) investigate the contradiction between short-run and long-run movements in the gold price and find that the gold price rises over time at the general rate of inflation and hence is an effective hedge against inflation under a set of conditions. Levin and Wright (2006) examine the factors that contribute to the fluctuation of gold price with cointegration and VECM techniques over 1976-2005. Their findings are three folds. First, there is a long-run relationship between the price of gold and U.S. price level. Second, there is a positive relationship between changes in the gold price and changes in the U.S. inflation, U.S. inflation volatility, and credit risk, while there is a negative relationship between gold price movements and changes in the U.S. dollar trade weighted exchange rate and the gold lease rate. Third, in the major gold consuming countries such as Turkey, India, Indonesia, Saudi Arabia, and China gold acts effectively as a long term hedge against inflation.
The recent volatility in commodity prices has triggered a flurry of new research, including analysis of inflation-hedging properties. Overall, the research provides quite strong evidence that commodities provide effective short-run protection against inflation. Greer (2000) estimates that the correlation during the 12 month return of the unlevered Chase Physical Commodity Index and annual U.S. inflation was 0.23 between 1970 and 1999. Against the change in inflation, the correlation was 0.59. Erb and Harvey (2006) estimate that from 1969 to 2003, changes in the rate of U.S. inflation explained about 43 percent of the variance of returns of the Goldman Sachs commodity excess return index (GSCI), with higher inflation leading to higher GSCI returns. However, they stress that the results vary widely across individual commodities. Kat and Oomen (2007) suggest that there are a number of commodities that provide an effective hedge against unexpected U.S. inflation, based on 12 month returns.
Some studies have taken a longer-run perspective. Gorton and Rouwenhorst (2006) find that correlations between commodity futures returns and inflation tend to rise and become statistically significant as the horizon stretches. Adams, Füss, and Kaiser (2008) also conclude that correlations between commodities, measured using GSCI excess returns, and U.S. inflation rises with the investment horizon, although these positive correlations do not hold consistently for inflation in the euro area and Asia. Worthington and Pahlavlani (2007) present evidence of the long-run hedging properties of gold based on a positive long-run relationship between gold and U.S. inflation in the post-war period.
Gold is often seen as an effective way for investors to hedge against the rise in prices of goods and services because its price moves in the same direction as inflation: when inflation rises, the price of gold rises and vice versa.
In the literature, there are two main streams of research devoted to studying the relationship between gold and inflation. The first assumes that inflation is one of the fundamental determinants of gold prices, while the second assumes only the relationship between inflation and gold prices. The concerned markets are London and New York. To our knowledge, only Harmston (1988) studied the French case. Inflation, one of the fundamental determinants of the gold price
The article of Lipschitz and Otani (1977), published by the International Monetary Fund (IMF), is considered as the first study dedicated to the private market4 for gold where its price has been able to move freely since 1968. It is also the first study that sought to explain the determinants of the gold price when it was no longer regulated by the monetary rules of the Bretton Woods regime. With quarterly data from 1968 to 1974, the authors built econometric models explaining the supply and demand of gold. Among the mentioned factors, inflation appeared to be a variable that had significant impact on the demand for speculation and hoarding of gold (excluding industrial demand). In the model of the demand for gold, the estimated coefficient of the variable "inflation" was positive. This means that an increase in inflation causes an increase in the demand for gold and so in the price of gold.
Sherman (1983) also proposed a gold price model with annual data over the period from 1971 to 1982. Among the determinants of gold prices, there were two factors concerning inflation: the expected inflation (measured by the excess of liquidity in the money supply) and the unexpected inflation. The results showed that there was a positive relationship between gold prices and these two variables. Koutsoyiannis (1983) built a multifactor model to study the fundamental factors affecting the evolution in the price of gold. His model was based on the equilibrium relationship between the supply and demand of gold. The data were daily and covered the period from January 1980 to March 1981. The author found that there was a positive relationship between inflation and gold prices. Cai et al. (2001) studied the effect of macroeconomic announcements on the future price of gold in New York (Comex) from 1983 to 1997 with intraday data. The results showed that the announcements about inflation in the US and Japan had significant impacts on the price of gold. Ghosh et al. (2002) used the cointegration test to investigate the relationship between the monthly gold prices and macroeconomic variables on the London market from 1976 to 1999.
This model confirmed the positive relationship between gold prices and inflation. However, the authors emphasized that this was not the case for the period from 1982 to 1995 where there was a divergence in the evolution in these two variables. Lawrence (2003) used quarterly data from 1975 to 2001 on the gold market in London. Through the correlation coefficients and the cointegration test, the author concluded that there was no significant relationship between gold prices and inflation. Levin and Wright (2006) studied the price of gold in London over the period from 1968 to 2005 with monthly data. The results of the multifactor regression model indicated that there was a positive relationship between gold prices and US inflation. However, the authors found that there was no significant relationship between gold prices and global inflation. This can be explained by the fact that gold is priced in US dollars on the London market. Lucey and Tully (2007) used the APGARCH model (Asymmetric Power GARCH) to examine the price of gold. On monthly data from 1984 to 2003, the authors found that there was no significant relationship between gold prices and inflation. Faugere and Van Erlach (2008) studied the determinants of gold prices by the "Required Yield Theory" model that they proposed in 2003. This model was built on the assumption that gold functions only as a safe haven. The data used were quarterly and covered the period from 1979 to 2002. The authors found that there was a positive relationship between gold prices and inflation. Artigas (2010) examined the impact of the money supply on the price of gold over the period from 2001 to 2009 with monthly data. The author used the spot price of the gold ounce traded in New York to build a multifactor linear model. The dependent variable was the gold price. The independent variables were the changes in the money supply (lagged to six months) in the United States, the eurozone, and India and Turkey. The results showed that there was a positive relationship between the money supply and the price of gold. This means that the rise in the money supply in the world (higher inflation) led to higher gold prices. The author also confirmed that the price of gold could be used as an indicator of future inflation.Mani and Vuyyuri (2011) studied the fundamental determinants of the gold price in India in the periods 1978-1979 and 1999-2000. In this gold price model, the authors included the following independent variables: expected inflation, expected interest rates, foreign exchange rate, stock market performance, price of silver, lagged gold price and a dummy variable (equal to 1 when significant events affect gold prices and 0 for other years). The results showed that the expected inflation coefficient was significant, which was explained by the important contribution of inflation to the US dollar/Indian rupee exchange rate. Thus, when inflation in India rises, the exchange rate between the dollar and rupee increases, the gold demand increases, and thus the gold price rises.
The relationship between gold price and inflation
The articles presented in this section are not dedicated to studying the gold price determinants, but focus only on the relationship between gold price and inflation. Jastram's book (1977) is regarded as the most comprehensive study on the relationship between gold prices and inflation. The author examined the annual change in gold prices in the UK from 1560 to 1976 and in the US from 1808 to 1976. As the name of the book, "The Golden Constant", implies, the author demonstrated the constancy of the real value of gold over the centuries. In terms of gold, the price of bread or a brick was the same in 1560 as in 1960. Jastram's results also showed that gold tended to lose its purchasing power during periods of inflation, and vice versa during periods of deflation. However, the price of goods always came down to the same level in terms of gold (Retrieval Phenomenon). In 2009, the book was reissued with two new chapters written by Leyland to extend the study until 2007. Leyland confirmed that the purchasing power of gold continued to be maintained from 1977 to 2007 in the UK and the US. Feldstein (1983) studied the relationship between inflation and the price of gold using the model of portfolio equilibrium. In contrast to the traditional theoretical conclusion that relative prices are unaffected by the rate of inflation, Feldstein (1983) showed that, because of unindexed taxes on capital income, a higher expected rate of inflation raises the prices of land and gold relative to the general price level of produced goods. More generally, a change in the expected equilibrium rate of inflation alters the real net rate of interest, the stock market value of real capital, and the real net marginal product of investment. In an economy with capital income taxes, inflation is far from neutral. The author concluded that changes in expected inflation can have powerful effects on the relative prices of such investment assets. Harmston (1998) took up the idea of Jastram (1977) and studied the changes in the annual purchasing power of gold in the US (from 1796 to 1997), the UK (from 1596 to 1997), France (from 1820 to 1997), Germany (from 1873 to 1997) and Japan (from 1880 to 1997). The results showed that gold maintained its long-term purchasing power in these countries, despite some periods of instability. Similarly, the purchasing power of gold increased after the end of the convertibility between gold and the US dollar in August 1971. However, gold tended to lose its purchasing power during periods of war or social and economic instability. Brown (1987) studied the futures market for gold in New York (COMEX) for the period from 1975 to 1983 with monthly data. The author concluded that futures on gold did not allow investors to hedge against inflation because their returns did not evolve with the same rhythm as inflation.
Laurent (1994) examined the role of gold in the international monetary system after the collapse of the Bretton Woods regime in April 1978. Even when gold was no longer a currency, it still played an important role in the monetary system as it could be used as an indicator of inflation because of its positive correlation with inflation. However, Lawrence (1994) showed that small changes in inflation were not always reflected in the price of gold. Garner (1995) argued that the price of gold can be used as an indicator of inflation. Despite its demonetization, it is always a store of value. When investors expect higher inflation in the future, the demand for gold (as a store of value) increases. Thus, the gold price increases. For this reason, an increase in the price of gold may be a sign of a future rise in inflation. Garner (1995) noted, however, that this indicator is not always reliable because there are also other factors in the demand for gold. Mahdavi and Zhou (1997) compared gold with other commodities (traded in London) for their effectiveness as an indicator of future inflation. The data used were quarterly and covered the period from 1958 to 1994. Using the technique of cointegration, the authors found that the price of goods was a better indicator of future inflation than gold.
Adrangi et al. (2003) used the monthly prices of gold quoted in London over the period from 1968 to 1999. To investigate the relationship between gold prices and inflation, the authors also used the cointegration test. The results show that there is a positive relationship between gold prices and inflation. Capie et al. (2005) investigated the role of gold in the hedge against the devaluation of the exchange rates between the US dollar and both the pound sterling and the Japanese yen. The authors used weekly data from January 1971 to February 2004. The methods used were cross-correlation coefficients, multivariate linear regression and the models of ARCH, GARCH and EGARCH. The results showed that there was a negative relationship between the price of gold in the US dollar and its exchange rates with the pound sterling and the yen. Thus, gold acted as a hedge against the devaluation of the dollar exchange rates. However, this role was not stable over time. It depended on unpredictable political attitudes and events. Ranson (2005) compared the effectiveness of gold and the inflation-linked bonds, called TIPs (Treasury Inflation Protected Securities) in the hedge against inflation. Several results emerged from the empirical study. First, the price of gold was a good indicator of future inflation. Second, gold was more effective in forecasting inflation than the consumer price index. Third, gold was more sensitive to inflation than TIPs bonds (a 1% rise in inflation caused a rise of 8.8% in the gold returns and a decrease of 2.8% in the TIPs bonds). From these results, he concluded that gold was a better hedge against inflation than the bonds indexed to inflation. In addition, Ranson (2005) showed that gold was a better indicator of inflation than oil because it was more correlated with inflation than oil. Pahlavan and Worthington (2006) examined the role of gold in the protection against inflation with monthly data from 1945 to 2006. The cointegration results showed that there was a positive long-term relationship between gold prices and inflation. The authors also concluded that investments in gold, both physical and paper, allowed investors to protect themselves against inflation.
McCown and Zimmerman (2006) also used cointegration tests to study the relationship between gold prices and inflation. The data included the price of gold in London (in US dollars) and the US consumer price index. The periodicity of the data was annual, quarterly and monthly and the studied period was from 1970 to 2003. The results showed that there was a long-term relationship between gold prices and inflation.
Dempster and Artigas (2009) used a monthly US database over the period from 1974 to 2008. The authors demonstrated that there was a strong positive correlation between gold prices and inflation during periods of high inflation (above 5% per year). They also found that real returns (net of inflation) of gold were higher than that of stocks, real estate and inflation indexed bonds (TIPs). During periods of high inflation, gold was less volatile than the other three assets. The authors thus concluded that gold was a good hedge against inflation.
Blose (2010) studied the relationship between gold prices and expected inflation. The original point of this article was that the author took into account the impact of expected inflation on the carrying cost of gold. According to the author, there are two possibilities. First, expected inflation causes an immediate change in gold prices. Second, expected inflation has no impact on gold prices. The explanations of the latter possibility are as follows. The expected inflation causes a rise of the interest rate (considered as a risk free asset). Then, the rise of the interest rate causes a rise in the carrying cost of gold investment. The rise in the carrying cost will cancel the speculative profit from investing in gold over the inflationary period (under the hypothesis that gold has a zero-beta in the CAPM model). Thus, the changes in the expected inflation have no impact on the price of gold. Using nonlinear regression on American monthly data from March 1988 to February 2008, the author found that the second possibility was validated. This means that expected inflation had no impact on gold prices.
Joy (2011) studied the relationship between gold and the US dollar in a weekly database from January 1986 to August 2008. The data included gold prices from the London market and 16 US dollar exchange rates (expressed in terms of home currency per dollar). The author used the conditional correlation with the multivariate GARCH model to study the relationship between gold prices and exchange rates. The results showed that the conditional correlation between changes in the price of gold and changes in the US dollar exchange rate was negative. This means that increases in the price of gold tended to be associated with decreases in the value of the US dollar. However, this relationship was not constant over time. Wang et al. (2011) compared the role of gold in the protection against inflation in the US and Japan with monthly data from January 1971 to January 2010. The authors studied the relationship between gold prices (in dollars and yen) and the consumer price index (in dollars and yen) in both the short-run and long-run. For the long-run relationship, the authors used cointegration tests (Engle and Granger 1987, and Enders and Siklos 2001). The advantage of the test of Enders and Siklos is that it allows verifying the non-linear cointegration relationship between two variables. For the short-run relationship, the authors tested the symmetric relationship between variables. If the short-run relationship is not linear and the long-run relationship is linear and stable (symmetric), the threshold vector error correction model is used to analyze the short-run adjustment process. If the long-run relationship is not linear and stable (asymmetric), the threshold cointegration-threshold vector error correction model is used. Finally, the causality test was used to analyze the relationship in the short-run between gold returns and inflation. The results showed that gold was only partially effective in hedging against inflation in Japan in the long-run. In the short-run, in the periods of low momentum regimes, gold was unable to hedge against inflation, in both the US and Japan. In contrast, in the periods of high momentum regimes, gold was able to hedge against inflation in the US but not in Japan.
2.3 Hedging Ability of Gold against Exchange Rate Fluctuation
Generalizing beyond gold, Clements and Renee (2008) study the joint determination of the prices of "commodity currencies" and "currency commodities." Countries that are thought to have 'commodity currencies" include Australia, Canada, New Zealand, and several developing countries that are rich in natural resources. The value of commodity currencies is hypothesized to depend on commodity prices. But there is presumably bivariate causality in that the prices of currency commodities depend on the exchange rates of commodity-exporting countries. The authors find that currencies are less affected by commodity prices than commodity prices are affected by currencies. They conclude that commodity-currency research failing to account for endogeneity between currency and commodity returns may be mis-specified. Sari et al. (2007) study the dynamic relations among the futures prices of oil, gold, silver, and cooper. They find a strong connection between gold and silver while copper appears to be nearly independent of movements in the prices of the other commodities. Gold and silver, help explain the volatility of forecast errors in oil prices. In addition, exchanges rates help explain the volatility in oil and metals. Since their analysis is focused on the volatility of forecast errors, their findings are particularly well suited for derivatives markets where volatility is a key valuation factor. Capie et al. (2005) apply an exponential generalized autoregressive conditional heteroskedasticity (EGARCH) technique to investigate the exchange rate hedge of gold price by using weekly data over the period 1971-2004. They find that the gold returns could be a hedge against U.S. dollar depreciation and that there is a negative relationship between gold price and sterling-dollar and Yen-dollar exchange rates but the strength of this relationship varied over time.
Sari et al. (2010) investigate the relations among precious metals, oil, and the US$/Euro exchange rate. They find no long-run equilibrium relations between commodity returns and changes in the exchange rate so they suggest diversification into the precious metals in the long run. However, they also find that precious metal prices and exchange rates are closely linked in the short run (2 days) and that aftershocks occur. Lizardo and Mollick (2010) investigate the relation between oil prices and the US$. They find that increases in real oil prices lead to a significant depreciation of the US$ against the currencies of oil exporter countries including Canada, Mexico, and Russia. On the other hand, an increase in oil prices brings depreciation relative to the US$ in the currencies of oil importers such as Japan. Moreover, when oil prices increase, the US$ tends to decline relative to currencies of countries that are neither net exporters nor significant importers, such as the UK.
Beckers and Soenen (1984) verify the gold/Dollar inverse relation empirically and draw a strikingly asymmetric implication for US and non-US investors. They say that the correlation between the Dollar price of gold and the Dollar's weakness implies that
…a non-US Dollar base currency investor will have to take into account his implicit foreign exchange risk position when he invests in gold or gold-linked instruments. It turns out that the total risk of his position valued in a non-US Dollar base currency is usually lower than when it is valued in US Dollars"
An early examination of these can be found in Carse et al. (1980); a recent study of hedging dollar risk using derivatives is Allayannis and Ofek (2001). With such choice available, why might gold be of interest? The answer is twofold. First, a range of products have developed which mean that one can in effect buy gold without actually taking possession of the physical commodity. Second, all the available techniques provide what can be termed a "special hedge". Protection is given least one currency fluctuated against some other specific currency. Gold, in contrast, is incapable of protecting against fluctuations of currencies in general. Whether it did so in practice is, of course, a different matter. That is the subject of the remainder of this paper.
CHAPTER THREE
RESEARCH METHODOLOGY
3.0 Simple Illustration of the Relationship between Gold Price and Inflation Rate
Long run relationship between gold and inflation in this paper is captured by:
where = log of gold price in RM
= constant or true value of gold price in RM
= hedging coefficient
= Inflation rate
= error term
The hedging coefficient denotes the hedging ability of gold against inflation. It also denotes the cross price elasticity between gold price and CPI. If the is 1, it means that percent change in CPI is equal to percent change in gold price. So, it can be said that gold is perfectly able to hedge against inflation. If the value is 0, it means that any percent change in CPI, will not affect gold price. If the value is between 0 and 1, it means that gold is able to hedge against inflation, but only partially.
3.1 Empirical Analysis
This paper will construct a nonlinear threshold vector autoregressive (TVAR) model for the empirical study. This paper will use the Malaysia's inflation rate fluctuation as the threshold variable to construct a high inflation regime and a low inflation regime. The empirical results will show that when the inflation rate increases by more than certain percentage, investing in gold could avoid losses from the inflation; otherwise, gold does not serve as a hedge against the inflation. We believe this finding could be a useful reference for Malaysia government and a guide for investors who would like to use gold as a hedge against inflation.
3.1.1 The model
The purpose of this paper to examine to what extent gold could serve as a hedge against inflation. To reach this goal, the model is specified as follows:
where is the inflation rate; is the gold price in RM, denotes the change rate of . When f ' > 0, this indicates that the gold return is large enough to cover the loss from the changes in the rising of inflation rate. Depending on the status of , equation (1) is as follows:
where is the threshold value of the inflation rate; therefore, could be used to divide the regimes in this threshold model.
3.2 Research methodology
The threshold autoregressive (TAR) model developed by Tong (1978) and Tong and Lim (1980) uses an optimal threshold value to divide the short-run dynamic status of one economic indicator into two regimes. When there are multiple regimes, the TAR model could be transformed into a TVAR model.
This paper must examine the existence of the threshold effect before estimating the TVAR model. We follow the approach of Tsay (1998) to test the linearity of the model. The null hypothesis is that the model is linear, this paper will use the VAR model and the alternative hypothesis is that the model is nonlinear then this paper will use the TVAR model. Tsay (1998) employs the recursive least squares method (RLS) to obtain the predictive residual of the arranged autoregression (ARR) to build the test statistic based on the standardized predictive residual. If the null hypothesis is rejected, which indicates that the model is nonlinear, then the next step is to find the values of the two parameters, the delay parameter and the threshold value.
The process of threshold value estimation
The threshold values are estimated with the threshold autoregressive
models developed by Tong (1978) and Tong and Lim
(1980). They use an optimal threshold value to divide the short-run
dynamic status of one economic indicator into two regimes. When
there are multiple (two) regimes, the threshold model could be
transformed as:
Î-t = A1 +Φ1;iÎ-t−i
Iðqt−d N γÞ+ A2 +Φ2;iÎ-t−i
ð1−Iðqt−d N γÞÞ+ε;
ðA1Þ
Î-t = xt
yt
21
; Α1 = α10
β10
21
; A2 = α20
β20
21
;
Φ1=
α1;11…α1;1p;α1;21…α1;2p
β1;11…β1;1p; β1;21…β1;2p
22p
Φ2=
α2;11…α2;1p;α2;21…α2;2p
β2;11…β2;1p; β2;21…β2;2p
22p
where Î-t is matrix of variables, Α1, A2, Φ1, and Φ2 are matrixes of
coefficients. p is the lag length; qt−d is the threshold variable, and d is the delay parameter; γ is the threshold value; and the error term ε
has the properties such that ε=(ε1 * ε2 * )′~iid, E(εt|Ωt−1)=0, and E
(ε t
2|Ωt−1)=σ2 where Ωt−1 is the information set in period t−1; I(â-¡)
are the indicator functions of regimes, and it is assumed that I
(qt−dNγ)=1 if there exist regimes and I(qt−d≤γ)=0 otherwise.
We must examine the existence of the threshold effect in Eq. (A1)
before estimating the threshold model. We follow the approach of
Tsay (1998) to test the linearity of the model. The null hypothesis is
that the model is a linear model-and the alternative hypothesis is
that the model is a non-linear model. Tsay (1998) employs the
recursive least squares method (RLS) to obtain the predictive residual
of the arranged auto-regression (ARR) to build the test statistic based
on the standardized predictive residual. For detailed discussion of the
Tsay linearity test, please refer to Tsay (1998).
If the null hypothesis is rejected, which indicates that the model is
non-linear, then the next step is to find the values of the two
parameters, the delay parameter d and the threshold value γ. The
threshold variable zt−d determines the appearance of the model in
two regimes: yt =
X′tΦ1 +Σ1=2
1 at If zt−d N γ
X′tΦ2 +Σ1=2
2 at If zt−d ≤ γ:
8<
:
ðA2Þ
If γ and d are given, then the earlier equation can be viewed as
having two independent linear regressive models, where Φi and Σ
are obtained as follows:
ˆΦ
i γ; d ð Þ = Σ
ið Þ
t
XtX′t
!−1
Σ i
ð Þ
t
Xty′t
!
; ˆΣ iðγ; dÞ
= Σ
ið Þ
t
yt−X′t ˆ φ
i
Þ yt−X′t ˆ φ
i
Þ′ = nið −kÞ
ðA3Þ
where Φ
i =Φˆ iðγ; dÞ; ni denotes the observations in regime i, i=1, 2;
and k indicates the dimension of Xt and kbn. The residual sum of
squares is:
Sðγ; dÞ = S1ðγ; dÞ + S2ðγ; dÞ; Siðγ; dÞ = trace ni−k ð Þ Ë†Î£ iðγ; dÞ
h i
; ðA4Þ
where γ and d are obtained from the following equation:
arg minγ;d Sðγ; dÞ; 1≤d ≤d0 andγ∈ R0:
After attaining the optimal threshold value (γ) and the delay
parameter (d), the best fit threshold model can be built.
Suppose that those parameters and the regimes are known. The threshold variable determines the appearance of the model in two regimes and after attaining the optimal threshold value (γ) and the delay parameter (d), the best fit TVAR model can be built. In our case of the gold return and inflation rate, the equation can be written as follows:
where α and β are parameters; and εit and ε2t indicate the error terms in the two different regimes, respectively. When , this indicates that the fluctuation of the inflation is larger than the threshold value, called as "regime 1"; otherwise, called "regime 2." This paper employed the Wald coefficient test to examine the causality between the variables. If the null hypothesis holds, it indicates that inflation rate does not Granger cause changes in the price of gold. The rejection of this null hypothesis means that inflation rate fluctuations do Granger cause gold returns changes. In addition to directly testing the hypothesis, observing coefficients of the lag variables could also provide indication of the causality between i and Δg. When the null hypothesis is rejected and the coefficient sum is positive, this indicates that gold could be an effective hedge against exchange rate fluctuation. Besides the causality between i and Δg, this can also examine the reverse causality between Δg and i. If the null holds, it means that Δg does not Granger cause i. The rejection of this null hypothesis indicates that the gold return does not Granger cause exchange rate fluctuation. Based on the coefficients of the lag, this paper can ascertain the direction of the short-term effects of the gold return on inflation rate in regime 1 regime 2. With the results of the causality test and the lag parameter analyses, this paper then can examine whether investing in gold can avoid the changes in the rising of inflation.
3.2.1 Unit root test
Nelson and Plosser (1982) indicate that many macroeconomic variables have a nonstationary characteristic. Since nonstationary variables do not fit the traditional requirement of regression analyses, directly using them in an ordinary least squares (OLS) estimation would cause OLS estimators to have biased asymptotic distributions and lead to the so-called spurious regression problem. The purpose of the unit root test is to determine the integration orders of time series so that we can know whether we have to difference a certain time series. This paper utilizes two methods to conduct the unit root test: the Augmented Dickey-Fuller test (Dickey and Fuller, 1979) and the Philip Perron. The constant term and the time trend are added when the test is being conducted.
3.2.2 Linearity test
Before constructing the TVAR model, this paper has to make sure whether there exists a nonlinear relationship among the variables. This paper employed the linear test suggested in Tsay (1998) and use as the threshold variable. The first step to conduct the test is to decide the optimum lagged periods. Akaike information criterion (AIC) and the Schwarz information criterion (SC) will be used to determine the optimum lagged periods.
If the relationship shows non-linearity, this paper will use as the threshold variable and estimate the TVAR model. Regime 1 corresponds to the situation in which the inflation rate is higher. If the threshold value is positive, it means that regime 1 represents a high depreciation period. On the other hand, regime 2 corresponds to a situation in which the inflation rate is smaller. If one simply uses the actual appreciation and depreciation rates to specify the two regimes, rather than let the model endogenously determine the threshold value to specify the regimes, then the estimation results will be biased. This is the major reason for this paper to use the more complicated nonlinear TVAR model. To make sure that the estimation results of the TVAR model would not violate basic statistic assumptions, we examined the autocorrelation and the cross-correlation of the residuals in the two regimes.
3.3 Data Collection
This paper will use the monthly data of gold price per ounce in RM based on the London PM Fix, RM/USD exchange rate and monthly consumer price index for Malaysia. The gold price data come from the World Gold Council Value Research & Statistics Database and the inflation rate data are obtained from the World Bank database. Our sample period is from January 1970 to September 2011, which gives us 501 observations. The 1971 was picked as it was the year when the Bretton Wood system collapsed and the convertibility of currency to gold was terminated.
RESULTS AND DISCUSSIONS
RESULTS AND DISCUSSION (Inflation rate)
Table 1 shows that the gold price and the inflation rate are both I (1) variables; that is, the two variables will be stationary after being first-order differenced.
Table 1: Results for Unit Root Test
ADF
Philip-Perron
Level
Intercept
Intercept and Trend
Intercept
Intercept and Trend
Gold price
5.418482
4.155139
5.602787
4.338961
Inflation rate
0.744621
-2.278771
1.019309
-2.083911
First Difference
Gold price
-15.61364***
-16.18977***
-18.91048***
-19.22886***
Inflation rate
-16.34718***
-16.37105***
-16.22247***
-16.22785***
Based on the results in Table 2, we can proceed to use lag 6 in the Tsay (1998) linearity test as out of 6 criterions, 4 showed significance at 10%.
Table 2: Lag Selection
Lags(p)
LogL
LR
FPE
AIC
SC
HQ
0
-1862.605
NA
6.609876
7.564319
7.581360
7.571010
1
-1815.688
93.26258
5.553679
7.390214
7.441336*
7.410287
2
-1808.269
14.68775
5.477183
7.376344
7.461547
7.409797*
3
-1801.375
13.59164
5.413275
7.364605
7.483889
7.411440
4
-1797.432
7.741173
5.414560
7.364838
7.518203
7.425054
5
-1795.420
3.935107
5.458431
7.372901
7.560347
7.446499
6
-1788.453
13.56615*
5.393181*
7.360865*
7.582393
7.447844
7
-1784.714
7.250487
5.398965
7.361924
7.617533
7.462285
8
-1780.472
8.191409
5.393760
7.360942
7.650632
7.474684
Table 3: Linearity Test
p
d
1
2
3
4
5
6
1
0.6807
0.5757
0.0863
0.0036
0.0004
0.0019
2
0.5273
0.4035
0.1225
0.0093
0.0097
3
0.0910
0.0045
0.0019
0.0009***
4
0.0105
0.0028
0.0220
5
0.0021
0.0032
6
0.2527
Applying this result in the linearity test, we find that when p=6 and d=3, the null hypothesis of linearity is rejected. The test result is reported in Table 3.
Threshold of inflation rate and inflation
Lag criteria for Causality test
∆ of gold price
Inflation rate
Inflation (upper)
Inflation (lower)
Inflation (upper)
Inflation (lower)
1
5.986539
5.992538
1.416754
1.527801
2
5.970289
5.982225
1.440261
1.546567
3
5.963857
5.987279
1.455195
1.569956
4
5.972239
5.997280
1.468418
1.591319
Table 4: Estimation Results
Dependent Variable
Null hypothesis
TVAR model (regime 1)
TVAR model (regime 2)
>10.91%
<10.91%
Sum of coefficients
Chi-square test
Sum of coefficients
Chi-square test
∆ of gold price
Inflation rate effect ∆ of gold price
0.132207
0.43503*
1.208773
0.563212*
Inflation rate
∆ of gold price effect inflation rate
0.005172
1.390866
0.008263
