Two PSA surveys, one difference
What a backyard farm is paid, against what a commercial one is paid
PSA surveys both populations every month and has done since January 2010. It publishes both figures and it never subtracts one from the other. Across 150 months where both surveys carry a figure, backyard farms were paid an average of ₱4.12/kg less than commercial ones — and were on the lower side in 123 of them.
Worth being careful about what that means. These are two surveys of two different populations, not one price measured twice. A backyard sale is often a whole pig to a neighbour or a local buyer at the farm; a commercial sale is a contracted lot moving to a processor on terms nobody publishes. So this is the size of the gap, not an explanation of it — and ₱4.12/kg on a 95 kg pig is about ₱391 a head, which is small against the ₱26.51/kg the gap reached in 2022-07 and not nothing.
01 · The gap, in numbers
Six figures, all subtraction
Every one of these is computed from the two surveys during the build. None of them is a figure PSA publishes, and none of them is typed here.
Latest
₱2.10
2026-06 — backyard above commercial.
Average, whole series
₱-4.12
Across 150 months where both surveys carry a figure.
Average, last two years
₱1.99
24 months. The gap has narrowed and changed sign.
Widest
₱26.51
2022-07, with backyard above commercial.
Months on the usual side
123
Backyard below commercial, of 150.
Months the other way
27
Most recently 2026-06 at ₱2.10/kg.
02 · The two prices
Both surveys on one scale
One vertical axis across both lines. Two charts each fitted to their own range are two different scales, and the gap between them would be invisible in each.
Farmgate price, ₱/kg liveweight
Backyard (household survey) · Commercial (establishment survey)
Source: the household survey and the establishment survey, live rows only. The backyard line breaks at the 1 month PSA published nothing; 49 commercial months carry no figure.
The difference
Backyard minus commercial, ₱/kg. Above zero means backyard was paid more.
Source: the derived spread series, recomputed from its two operands on every run.
03 · How this is computed
The arithmetic, and where it is checked
Every row here is hog-farmgate-household minus hog-farmgate-establishment for the same month, rounded to 2 decimal places, with both operands written onto the row. A month where either side has no figure has no difference, and is null rather than a subtraction from a blank.
Every row of the gap series carries that arithmetic in a machine-readable form, and this archive's checker resolves both operands and recomputes the subtraction on every run. A derived figure that stopped matching its inputs fails the build rather than sitting on a page being wrong.
04 · Check it yourself
The last two years, all three columns
Both operands beside the difference, so the subtraction is checkable by eye. A derived series whose arithmetic a reader cannot follow is asking for exactly the trust this archive is built to avoid asking for.
| Month | Backyard | Commercial | Difference |
|---|---|---|---|
| 2026-07 | — | — | — |
| 2026-06 | 172.62 | 170.52 | 2.10 |
| 2026-05 | 171.78 | 171.25 | 0.53 |
| 2026-04 | 172.24 | 176.55 | -4.31 |
| 2026-03 | 181.33 | 176.23 | 5.10 |
| 2026-02 | 181.36 | 169.10 | 12.26 |
| 2026-01 | 175.16 | 168.85 | 6.31 |
| 2025-12 | 185.28 | 189.92 | -4.64 |
| 2025-11 | 176.13 | 180.45 | -4.32 |
| 2025-10 | 184.75 | 181.88 | 2.87 |
| 2025-09 | 192.17 | 185.23 | 6.94 |
| 2025-08 | 192.40 | 189.32 | 3.08 |
| 2025-07 | 195.53 | 191.07 | 4.46 |
| 2025-06 | 212.12 | 217.66 | -5.54 |
| 2025-05 | 208.52 | 213.99 | -5.47 |
| 2025-04 | 208.49 | 212.70 | -4.21 |
| 2025-03 | 212.03 | 215.77 | -3.74 |
| 2025-02 | 210.78 | 214.63 | -3.85 |
| 2025-01 | 201.45 | 207.34 | -5.89 |
| 2024-12 | 186.86 | 178.56 | 8.30 |
| 2024-11 | 179.37 | 177.72 | 1.65 |
| 2024-10 | 179.47 | 171.28 | 8.19 |
| 2024-09 | 190.43 | 170.78 | 19.65 |
| 2024-08 | 177.22 | 168.90 | 8.32 |
A dash is a month one of the surveys did not publish. There is no difference for such a month, because a subtraction from a blank is not a smaller number — it is no number.
Both surveys also exist per province — by region and province — where the range between the dearest and cheapest region is several times the size of this gap.