Yield and price: how a UK harvest shock reaches the market

CropIntel forecasts UK wheat yield. Price is what you ultimately trade. This page sets out, honestly, how the two connect over the years we have ex-farm price data for, and, just as importantly, where the link is weak and why.

The short version
A smaller UK harvest tends to firm domestic ex-farm price; a big harvest tends to soften it. That direction holds in the data. But the relationship is loose: the UK is a price-taker, and global supply, demand, and currency move flat price far more than the British harvest does. The durable value of a UK yield edge shows up in basis, regional premium, and the timing of risk, not in calling flat price.

The economic logic

Wheat is fungible and traded globally. A poor UK harvest tightens the domestic balance sheet, less home-grown grain, more import pull, firmer regional premiums over the futures benchmark. So the sign of the relationship is exactly what theory predicts: lower yield, firmer price. What theory also predicts, and the numbers confirm, is that the magnitude is small relative to the global drivers stacked on top.

Year by year

Our national directional lean, the actual UK yield anomaly (versus trend), and how ex-farm feed wheat moved across that harvest's marketing year (post-harvest autumn → the following summer). Price data begins in 2015. The lean is shown only where it was strong enough to stand behind; the confidence-filtered, region-by-region accuracy that gives the 62.3% headline is on the track record.

Marketing year Our yield lean UK yield anomaly Ex-farm price move
2015/16Near avg+0.11+9.2%
2016/17Near avg+0.88+10.5%
2017/18Near avg+0.05+22.7%
2018/19Near avg+2.04-20.2%
2019/20Below avg-1.66+23.4%
2020/21Near avg-0.05+6.4%
2022/23Near avg+0.48-30.9%
2023/24Below avg-1.03-3.4%
2024/25Near avg-1.41-10.0%

How tight is the link? Honestly: loose.

Across the 9 marketing years with price data, the correlation between the UK yield anomaly and the ex-farm price move over the year is r = -0.38 (p = 0.32, n = 9). The negative sign is the right direction, lower yield, firmer price, but it is weak and not statistically significant on this sample. Yield explains only a small share of the year-to-year price swing.

The clearest example of the limit: 2022/23. The UK harvest came in +0.48 versus trend (slightly above normal, which all else equal should soften price), yet ex-farm feed wheat moved -30.9% across that marketing year. That swing was the global market unwinding the 2022 invasion-spike, nothing to do with the size of the British harvest. When the world moves, UK yield is a rounding error on flat price.

Why we don't sell this as a price forecast
Claiming we predict UK wheat price from yield would be overreaching, the data doesn't support it, and a buyer who re-ran the numbers would rightly discount everything else. We forecast the UK supply shock (yield), and we're transparent that flat price is a global variable. That honesty is the point.

Where a UK yield edge actually pays

Knowing the UK harvest before the trade reports do is valuable, just not as a flat-price punt:

Our skill is at the yield step

This page is about the downstream link from yield to price, which is loose. The thing CropIntel is actually good at is the step before it, calling the UK yield itself, with a walk-forward hit rate of 62.3% over 23 years. That record, region by region and year by year, is on the track record.

Related: Today's call · Track record · By the numbers · Methodology. Not financial advice, see the footer.