Building an efficient and fair commodities marketplace requires collaboration, and much less of the over-used term disruption
High-volume businesses dealing in homogenous commodities with stable supply routes can rely on traditional broker-led trading to achieve the best price. But what about the other half of the commodities sector, whose products trade in smaller, more volatile differentiated segments, and for whom an irregular supply chain is a fact of life?
The answer is to facilitate smaller, more agile, location-specific exchanges where goods are bought and sold in a digital market to achieve their optimal price.
This blog post discusses how Perfect Channel worked with Global Dairy Trade (GDT) to create ‘GDT Marketplace’, built on Perfect Channel’s auction and trading platform Perfect Exchange. We have elaborated on the methodology in our white paper: ‘The auction mechanisms behind delivering a best in class dairy trading platform’.
Addressing GDT’s need for an online marketplace
An auction design that works successfully for one industry will not necessarily be the best suited for another industry. When Global Dairy Trade (GDT) began the process of looking for a technology company they had the following requirements:
- A technology company that could help grow their business of matching buyers and sellers in the global dairy industry.
- A partner who could help them create a full feature, configurable marketplace platform that combines enterprise grade performance.
- A partner who could offer different trading models for their customers to select according to their individual preference.
- A technology partner with a deep understanding of auction trading mechanisms.
GDT’s Trading Model
GDT Marketplace has initially offered two types of trading model to its participants: Fixed Price and Tender. Universally, the Fixed Price model, is where the lister specifies the sale price (visible only to the buyers they have approved for that listing), and the product is sold to buyers in the order in which they submit purchase requests until the available quantity is sold. In the Tender model (which is the focus of this blog post), both the sale price and the allocation of quantity between buyers is determined by a software algorithm.
A Collaborative Approach
Turning GDT’s (three) objectives into an algorithm that could resolve any Tender auction scenario presented to it, was the result of a close collaboration between GDT and Perfect Channel. Perfect Channel subsequently used their marketplace platform to work with GDT to address their three objectives as follows:
1. Maximise total value gain across buyers and sellers. This first objective reflects the importance for the trading platform to present genuine trading opportunities – not speculative listings at unrealistic prices – and to match the right buyers with the right sellers.
2. The second objective to simplify the decision processes faced by buyers reflects that in many auction designs the process for buyers to enter bids may initially appear very simple. GDT’s research found it is much better to keep the bidding decision as simple as possible through a structure where the buyer’s best strategy is to bid the price they are willing to pay without needing to take into account what other buyers are likely to do.
GDT operates in a trading environment where the available supply can be divided between multiple successful bidders, and where most buyers and sellers will be managing a flow of product through repeated transactions. In some cases, a buyer or a seller may have specific quantities they need to trade, but more often they will have some flexibility as to how much product they buy or sell through their current transaction versus their next trade.
3. As a result, the third objective was to enable buyers and sellers to reflect their degree of flexibility, whether very narrow and highly specific or very wide and unconstrained. Enabling buyers and sellers to signal their flexibility enables the auction system to find the best combination of winning buyers that achieves the highest overall value gain while ensuring no buyer or seller is allocated quantities that are unacceptable to them.
The result is a trading mechanism where all the sellers have to decide is the range of quantity they are willing to sell, and all the buyers have to decide is what quantities of product they would be willing to buy at various key price-points for them (i.e. they specify their demand curve). The Tender algorithm developed by GDT and Perfect Channel will automatically resolve the auction scenario (or ‘clear the market’) through aiming to maximise total marketplace value at a fair market price that benefits everyone.
This blog elaborates the challenges faced with respect to creating a specific marketplace (in this case dairy) and offers an insight on the best steps to address them. There is no prescriptive model, and certainly no ‘one size fits all’ solution that can be applied to all commodity markets. Core to every successful market design is a strong partnership between industry and the electronic trading venue. At Perfect Channel we believe that to build an efficient and fair commodities marketplace requires collaboration, and much less of the over-used term disruption.
For more content: