Identifying possible misspecification in South African soybean oil future contracts

Abstract

Soybean crushing plants operate on a crush margin, which is the monetary difference between the combined sales value of mainly soybean meal and soybean oil and the cost of raw soybeans. However, given the high volatility in the prices of these three products, crushing plants normally secure these prices simultaneously. If not, they are vulnerable to the relative price variation between these three products.

Futures markets, such as the Johannesburg Stock Exchange (JSE) Commodities Derivatives Market (CDM) (previously known, and hereafter referred to, as the South African Futures Exchange (SAFEX)), provide futures contracts that can be used as a mechanism for securing these prices. Soybean crushing plants would usually buy soybean futures contracts whilst simultaneously selling soybean meal and soybean oil futures contracts (in a ratio aligned with production), thereby securing the processing plant’s gross margin or better known in the industry as the ‘crush margin’. But this is only viable given adequate liquidity within these futures contracts (which is not the case for SAFEX soybean oil futures contracts). Furthermore, if South Africa is a net importer of the underlying commodity, as is the case with soybean oil, the CBOT contract, as traded on SAFEX futures’ price normally represents the majority of the import cost also known as the import parity cost. Therefore, with most soybean oil usually being imported from Argentina, one would expect SAFEX soybean oil futures contracts to reflect the cost of imported soybean oil from Argentina (which are significantly different at times through the season).

However, currently (2020), the SAFEX soybean oil futures contract is a CBOT contract, that is dual listed and cash-settled . The research study seeks to determine whether this is a misspecification and whether or not SAFEX soybean oil futures contracts should rather be based on the Argentina fob soybean oil prices which is a much better representation of South Africa’s import parity and local industry prices. If correct, it may also explain why market participants are reluctant to utilize SAFEX listed CBOT soybean oil futures contracts, explaining the low trading volumes and inadequate liquidity.

Hence, the study used the Engle-Granger (1987) cointegration approach, alongside a range of diagnostic tests to evaluate the existence of adequate long and short-run cointegration relationships amongst a linear combination of data variables underlying the current specifications of SAFEX soybean oil futures contracts versus that of an alternative linear combination of data variables that are cash settled of Argentina fob prices (settlement values). Essentially evaluating its efficiency under Eugene Fama’s semi-strong-form of market efficiency, in an attempt to identify possible misspecification by referencing CBOT settlement values as opposed to Argentina settlement values that could ultimately lead to greater participation and improved liquidity.

The study however failed to produce overwhelming statistical evidence for using Argentina settlement values as opposed to CBOT settlement values. Diagnostic tests revealed possible misspecification amongst the long-run equilibrium relationships for both CBOT and Argentinian soybean oil future prices, while concluding for no-misspecification amongst CBOT soybean oil future prices in the short-run. These results suggest that SAFEX soybean oil futures contracts does not incorporate all the information used by market participants in forming a prediction of subsequent spot market prices in the long-run. But does however incorporate sufficient information for such practices in the short-run, attracting speculators who hope to profit from short-term price variations in the absence of hedgers (typically soybean crushers) who in turn seek to employ effective long-term hedging strategies.

Therefore, the study rather pointed towards using CBOT settlement values until South Africa becomes self-sustainable, meeting local demand with local production. In such case, a local physically settled soybean oil futures contract should be listed that accurately reflects local supply and demand conditions, given the collective participation amongst the majority of market participants within the South African soybean industry.