BARLEY PRICING ANALYSIS
Introduction
Like all of the commodities on the price of feed and malting Barley changes over time. For example, plotted below is the bid price of Feed Barley on Hectare Trading.
Being able to inform sellers of the general price of barley in their area would be a valuable addition our offering on the Market Insights page, however our trading is much sparser than our Wheat bid information. This analysis will investigate the possibility of building a barley pricing prediction based on internal and external pricing data.
How do Hectare Trading prices compare to external market data?
The first step in analysing if we can predict barley prices is to ensure that the price is coupled to market prices. I.e. does barley pricing act in the same way on Hectare Trading as it does in the wider UK agricultural market?
Unfortunately, we don't have a trading market that we can compare our barley prices to, as with the london feed wheat prices published on ICE/LIFFE. However, AHDB does publish weekly ex-farm pricing information for feed barley in the corn returns dataset.
Let's compare the AHDB corn returns data to our Hectare Trading spot feed barley bid prices over time.
We can see that our feed barley bids are tightly coupled with the AHDB corn returns. Let's check how well these two prices are correlated.
Statistically, the two prices are moderately correlated (with an R number of
Does feed barley display the same regional carry as feed wheat?
When analysing regional carry in feed wheat pricing, we saw a progressively increasing carry from Southern to Northern regions of the UK. By comparing our feed barley prices to their respective AHDB prices per region we can plot the regional barley carry.
We can see that our barley regional biases are very similar to feed wheat.
| Region | Feed Carry | Barley Carry |
|---|---|---|
| South | −£4.55 | -£4.10 |
| Midlands & Wales | -£0.60 | -£0.80 |
| North | +£2.25 | +£8.30 |
| Scotland | +£10.50 | +£9.30 |
The only exception being the North, but this is most likely due to the low sample size. This means that we can assume that compare feed wheat and barley prices within regional groups.
How well do Hectare Trading feed barley and wheat prices correlate?
Since the regional carry for feed barley and wheat seem to be very similar, let's check to see how correlated these two prices are. If the two prices are highly correlated, we may be able to use the predicted prices drawn from the LIFFE to also predict barley prices per region.
Let's plot the regressions between each feed barley price and the average price of feed wheat in the super region on the day of the barley price:
Apart from the Midlands, there isn't a large enough sample size to calculate any statistically significant correlation between feed wheat and feed barley pricing.
However, we can see that in the Midlands there is a moderate correlation (with an R number of
Unfortunately, even this is quite a low sample size over a large amount of time and without enough samples in all regions it won't currently be viable to use the direct correlation between wheat and barley prices to give a daily prediction of feed barley prices on Hectare Trading.
Can we use the AHDB corn returns to create an barley/wheat offset?
We've seen that it's possible for us to calculate and apply milling wheat premium to our estimated feed prices using the AHDB corn returns data. Could we create a similar offset between feed wheat and barley and apply it to our predicted feed prices per region to obtain a daily regional feed barley price?
Firstly, let's compare the AHDB corn returns UK wide feed wheat and feed barley prices.
We can then take the difference between these prices over time. To get a 'Barley/Feed' offset of sorts
This weekly offset can then be applied to the predicted feed wheat price that is based on the London feed wheat LIFFE price over time.
Below see feed wheat bid prices in the Midlands over time, plotted against the offset feed wheat predicted price (original predicted feed wheat price also plotted for reference).
If we calculate the same error rate metrics as we did for the LIFFE feed wheat estimator for each region we can see that the a relatively accurate prediction of the feed barley price.
Note that we still do not have enough data in the North to validate this prediction, but using this method we could in theory still calculate a predicted barley price for the North.