آزمون ارتباط بین المللی در بازارهای آتی فلزات گروه پلاتینیوم

This study tests whether an international market exists in the platinum-group metal (PGM) futures markets. For this purpose, we tested the law of one price (LOP) and the causality between the U.S. and Japanese platinum and palladium futures markets. We also performed the test when structural breaks are considered. Long-run price relationships were found in both platinum and palladium markets but the LOP only sustained in the palladium market. The causality test revealed that it is the U.S. market that leads the price to transmit information between the U.S. and Japanese markets. Structural breaks had large impacts on the test results, suggesting that incorporating breaks is important when investigating the international price linkage in the PGM futures markets.

As vehicle emission standards become stringent worldwide, we see an increase in the demand for the platinum group metals (PGMs) from the auto industry. More than half of the platinum and palladium supplied internationally in 2008 was used for the catalytic converters (Johnson Matthey Plc, 2009). Unless technological improvements occur in the catalytic converter production and substitutive materials for platinum and palladium become available, demand for these PGMs will continue to rise in countries that have higher emission standards. If the differences in vehicle emission standards affect the demand structures of regional PGM markets the prices of PGMs may differ by their market locations. Hence, whether an international PGM market exists and how regional PGM markets are linked are becoming important issues for the participants in the PGM markets to obtain valuable price information. In theory, as stated by the law of one price (LOP), identical goods must have a single price (Lamont and Thaler, 2003) and if the world PGM market follows this law the regional PGM markets should be integrated as one market. Because PGMs are easier to standardize and store compared to other commodities it is likely that the price difference is small among regional PGM markets and that a world market exists for the PGMs. However, if transaction costs and trade barriers are high among the regional markets, these markets will not be integrated. Asplund and Friberg (2001) state that violation of the LOP occurs when demand and costs are different across locations and there are price rigidities and fluctuations in exchange rates. Therefore, if the demand structures are different among the regional PGM markets the LOP condition will be violated and there will not be a world market for this commodity. Identifying whether the LOP holds for the world PGM market will be valuable not only for the suppliers and consumers of PGMs but also for arbitrageurs and speculators trading the PGMs as financial asset. If the LOP holds among the regional PGM markets it will mean that in the long-run the regional PGM prices become the same and price information of the other markets will be useful even when the markets are far apart or traded at different time zones. Although there are several studies testing the LOP for different commodities such as wheat (Goodwin, 1992), soybean meal (Yang et al., 2000), fish (Asche et al., 1999 and Asche et al., 2004), lumber (Nanang, 2000), natural gas (Walls, 1994), and cars (Goldberg and Verboven, 2005), not many studies have tested this condition for the platinum and palladium markets. Previous studies testing the market linkage among different locations for the precious metal futures markets focus on the price relationship between the spot and futures prices (Chow, 2001) or on the difference in volatility (Xu and Fung, 2005). For study on the information flow between the U.S. and Japanese precious metal futures markets, Xu and Fung (2005) find that it is the U.S. market that plays the leading role in cross-border information transmission. A study testing the price linkage for the U.S. and Japanese gold and silver futures markets (Aruga and Managi, 2011) also reveals that information flow between the futures markets of the two countries are led by the U.S. market. Hence it could be that the information flow for the PGM futures markets of the two countries is also led by the U.S. market. The objective of this study is to provide some empirical evidence on whether an international market exists for the PGM markets by testing the LOP and the causality of price information flow between the U.S. and Japanese PGM futures markets. For this purpose, we use the platinum and palladium futures markets of the New York Mercantile Exchange (NYMEX) and the Tokyo Commodity Exchange (TOCOM), which are the world's two largest futures markets for PGMs. The LOP test will identify if the U.S. and Japanese PGM futures markets can be integrated as one market. In the causality test we will find the direction of information flow between the PGM futures markets of the two countries. This will allow us to see whether the U.S. and Japanese PGM futures markets are dependent. We will also test the LOP and causalities among the price series when structural breaks are considered because recently several studies suggest that the price linkage can be affected by structural breaks and that price relationships among the price series can become different before and after the break periods (Beyer et al., 2009, Aruga, 2011 and Aruga and Managi, 2011). Although many other related studies incorporate structural breaks in the cointegration methods (Hansen and Seo, 2002 and Park et al., 2007) we use the general cointegration method for testing the price linkage because we are more interested in identifying effects of structural breaks on the price linkage rather than testing the overall price linkage by including the breaks in the test model. This way of testing the price linkage for periods before and after the breaks allows us to see in detail how the breaks in the series affected the price linkage. If we use special treatments to incorporate the breaks for testing the price linkage in the cointegration model we will not be able to examine the changes in the price linkage before and after the break periods. There are still relatively a few studies focusing in the effects of structural breaks on the price linkage but Aruga and Managi (2011) have shown that the price linkage between the U.S. and Japanese gold and silver futures markets was influenced by the structural breaks in the price series. It is likely that if such structural breaks do exist in the PGM price series we might find that the price linkage between the U.S. and Japanese PGM markets is also affected from the structural breaks. Testing the effects of breaks on the price linkage is important because we use the 2001–2010 period in our study which includes adverse events such as the global financial crisis of 2008 and it can be that such events have influence on the price linkage for the U.S. and Japanese PGM futures markets. In the next section, we explain the methods used in this study. In the third section, we illustrate the details of the data. The results of the tests conducted in this study are described in the fourth section. Finally, the last section provides some concluding remarks.