What Drives Daily Volatility in Lumber Futures Markets?

Summary

The research studies the patterns of daily volatility in lumber futures contract prices traded at the Chicago Mercantile Exchange (CME) across different phases of U.S.-Canada softwood lumber trade dispute.

Situation

Lumber plays an important role in new residential construction. Because of its reliability and safety more than 90% of homes in the United States are built with lumber. However, lumber prices fluctuate over time, causing firms engaged in producing, processing, marketing, or using lumber and lumber products to face price risk. To help to manage this risk due to spot price volatility, the Chicago Mercantile Exchange (CME) launched lumber futures contracts in 1969. Futures markets provide information on future cash price expectations of traders on a daily basis and are found to be the primary center of price discovery for the underlying cash commodity. Thus, lumber futures prices are forecasts of future lumber cash prices, adjusted for storage costs and anticipated supply and demand conditions. However, lumber futures prices also exhibit volatility and volatility persistence. The process of market price reaction to information flows is argued to result in such persistence. But there are other reasons to expect volatility persistence in the case of futures contracts written on commodities. Aggregate physical inventories play a shock absorbing role in commodity markets, implying that when physical inventories are large, the size of commodity futures price changes is small. Further, aggregate inventories evolve slowly. Thus, commodity futures volatility is characterized by phases of varying length, depending on the speed of inventory changes for the particular commodity. A second factor influencing volatility in commodity futures is the time gap between when news arrives to the market and when the contract calls for delivery. Near-term information shocks should have greater impact on contracts for near-term delivery than on contracts for farther-out delivery due to the smaller elasticities of supply and demand for shorter runs. In the case of lumber, price volatility can be affected by supply factors, such as stumpage costs, railroad strikes or railcar shortages, and demand factors such as residential construction, interest rates, and prices of substitutes like steel or aluminum. Further, uncertainty due to softwood lumber trade disputes between U.S. and Canada is another source of volatile lumber prices. The softwood lumber trade dispute has been the largest and longest lasting dispute between the two countries. Canada, being the largest exporter of softwood lumber to the U.S., provides more than 90% of U.S. total imports. The main issue underlying the dispute has been whether the Canadian lumber is subsidized and if so, whether the U.S. lumber industry is adversely affected.

Response

This research studies the response of lumber futures prices to unobserved information flows. It is hypothesized that the price response to unobserved information flows depends on inventories and time to delivery. Further, the research compares daily lumber futures volatility across different periods of the U.S.-Canada softwood lumber dispute. Specifically, daily lumber futures prices from the Chicago Mercantile Exchange are analyzed from 1992 to 2005, defining volatility as the absolute value of log price changes over a day. Studying time-to-delivery effects requires using data on all contracts traded on a given day. This, in turn, requires a recognition of the correlation among price observations from the same day, which are subject to common shocks. A Generalized Least Squares (GLS) procedure is used to take this contemporaneous correlation into account, resulting in efficient use of futures price data and consistent standard errors of the estimates.

Impact

Daily lumber futures volatility is found to vary in different phases of the U.S.-Canada trade dispute. Daily volatility is higher in both the Softwood Lumber Agreement (SLA) period (1996-2000) and the post-SLA period (2001-2005) compared to the period with trade disputes and temporary tariffs (1992-1995). Also, volatility in the post-SLA period is higher than the volatility in the SLA period. This shows that uncertainty about future trade terms contributes to the price volatility. However, and most importantly, even the SLA period, created lumber price volatility rather than reducing it by eliminating uncertainty. The SLA period is characterized with a tariff-rate quota, creating a situation in which supply cannot respond immediately to a change in demand. An inverse relationship between inventory levels and lumber futures volatility is found, as predicted by the theory of storage. As inventory levels become smaller lumber futures contracts become more volatile. Moreover, the inventory effect on volatility varies across different periods of the trade dispute. A change in inventories from their minimum value of $3.1 billion to their maximum value of $7.5 billion causes a 3.9 percentage point decrease in daily volatility in the 1992-1995 period, whereas the same change in inventories causes a 0.9 and a 0.6 percentage point decreases in volatility in the 1996-2000 and 2001-2005 periods, respectively. An inverse relation between time to delivery and volatility is also found. The closer the contract trade date is to delivery time, the higher is price volatility.The results imply a 0.8 percentage point increase in volatility over the life of the contract. Compared to a typical day's absolute log price change of 1.2 percentage points, these changes are considered to be economically significant. This result can be interpreted as lumber supply and demand curves becoming more inelastic as time to delivery nears. Thus, shocks originating from either the supply or demand side of the market have a larger price impact as time to delivery nears. Further, it is found that while volatility persistence is statistically significant in the marginal distribution of lumber futures returns, much of that persistence can be explained by the dependence on time to delivery. The effects of these observable and exogenous variables on the price volatility of lumber futures may be exploitable to improve hedging, options pricing formulas, and the setting of margin requirements. Market participants, including lumber mills, lumber wholesalers, home builders, and construction companies, who are concerned about managing price risk should take into account the volatility determinants presented in this study. This study also has implications for forest management. Selling timber is a large component of income generated from a forest land. Even though the objectives for owning forest land might change over time, sooner or later most forest owners decide to harvest timber. Price the forest owners receive from their sale of standing timber depends on several factors, including the current state of the economy, market cycles, and lumber prices. The value of timber, in fact, changes frequently with changes in lumber prices. Any event that changes the demand for and supply of lumber, such as housing starts, tariffs, and other shifts in international trade conditions, causes fluctuation in lumber prices which in turn results in fluctuations in prices received for standing timber. Therefore, a forest owner should keep a close eye on lumber price volatility to determine the best time to sell timber.

State Issue

Other Issue

Details

• Year: 2011
• Geographic Scope: International
• County: Clarke
• Program Areas:
  • Agriculture & Natural Resources