Method
How a benchmark price is calculated, and how far each figure can be trusted.
Estimation
The benchmark price comes from a robust regression of value on quantity. Observations that do not fit are found with a statistical test on each one and left out, and large flows are protected. Each observation is weighted by its quantity, so the price equals total value divided by total quantity of the observations kept.
- For one trade route (an importing country and an origin), each month is one observation.
- For the whole EU, each member state is one observation, so one large country cannot drown the others.
- Changes over time compare the last 12 months with the 12 before, and the last 3 months with the same months a year earlier. For all origins together we use a Fisher price index, so a shift from cheap to expensive origins is not reported as a price change.
How reliable is it?
Every benchmark shows:
- the 95% interval of the price, and the fit (R²; below 0.7 means use with care);
- the observations used and left out, and the share of the trade value the figure rests on;
- normal variation: how much the price usually moves from month to month;
- agreement: in how many EU countries the price moved the same way;
- warnings when a product code mixes very different goods, or when the weight per piece shifts;
- a label: Reliable, Fairly reliable or Use with care.
Comparing your price
Your price is compared as you give it, with no adjustment. The benchmark is a price at the border, so a price delivered to your door naturally sits somewhat higher. Your price is placed among the market's monthly prices of the last 12 months. Your own change since a date you give is compared with the market's change over the same months.
Our rules
- Normal variation. With m the median month-to-month movement, a change between two periods of k months is within normal variation when it is smaller than 2 × m × √(2/k). When m is unknown, the band is ±5%.
- Reliability labels. Reliable: interval within ±5%, R² at least 0.7, at least 80% of the trade value used. Fairly reliable: within ±15% and R² at least 0.5. Use with care: everything else, and any benchmark from fewer than 3 observations.
- Price position. Within the typical range between P25 and P75 of the monthly prices; below it under P25; above it over P75; well above it over P90 or 1.5 × P75. With fewer than 12 months of data, or on a route with grade D, we say there is not enough data.
- Route grades A to D, from the last 12 months: A when the price varies less than 10%, with at least 24 months of data and €500,000 of imports; B under 20%, 18 months and €100,000; C under 40% and €20,000; D otherwise. A route with an implausible price per kg cannot get A or B.
- Large flows. Flows over 10% of the quantity are never set aside automatically. We measure the share on quantity, not value, so a mis-priced flow cannot protect itself.
- Minimum data for a change. At least 6 monthly figures in both periods for one route, or at least 3 countries pooled together.
- Supplier report. Each product is compared with the benchmark from all origins, and next to it with the one from its production country. The overall figure is weighted by volume when every product has one, otherwise a simple average. The report says which.
- Products far from the others. In a list, a product whose position is far from the rest is marked and left out of the average. It usually means a price in the wrong unit or a very special product. Never applied to a majority of the list.
- Per kg and per piece. Every product code has a benchmark per kg. A benchmark per piece is added where a stable volume reports that unit. The two are never mixed. Your price is compared in your own unit first.
Data we keep
When you prepare a supplier report, your prices, volumes and the supplier's name stay in your browser. To look up benchmarks, we receive only product codes, countries, currencies and months.
In the free version, we keep one anonymous line per compared product: the product code, the countries, the month, how your price compared in percent, a coarse volume band, and which benchmark it was compared with. We never keep who you are, who the supplier is or what you pay. No figure from these lines is ever shown with fewer than 5 contributions behind it. Premium access, when it opens, keeps nothing.