How to Read PSA Hog Farmgate Data Yourself

Every price article on the Philippine internet cites PSA, including ours. Almost none tells you how to open the table. It is one free page, no account, and about three minutes: OpenSTAT table 0032M4AFN12, commodity type "Hog for Slaughter", pick your province. That is the whole trick.

The rest of this is the part that takes longer to learn, which is why your own query will sometimes disagree with everyone else's and which of you is wrong.

The three-minute version

1. Go to openstat.psa.gov.ph and find Prices, then Farmgate Prices, then the Livestock and Poultry tables under the "NFG" database.
2. Open 0032M4AFN12, "Farmgate Prices of Livestock and Poultry, All Farm Types, by Geolocation, Commodity, Year and Period".
3. In the Type dimension pick Hog for Slaughter. There are other pig entries and picking the wrong one is the most common mistake.
4. In Geolocation pick your province. In Year and Period pick the last two years of months, not just the newest one.
5. Download as CSV or Excel. You now have the same rows every price article in the country is quoting at you.

If you would rather not do any of that, we already did it for all 110 geographies and put the result on our data pages as a CSV with the source URL and the date somebody actually opened it on every single row: the national all-farms series, the same thing by region and province, and the province grid if you would rather look at a picture. The file formats are documented so you can script against them.

What the series actually is

A farmgate price is what the farmer received at the first point of sale, net of the marketing costs the farmer paid. It is a producer price, not a market price and not a retail price. It is collected by survey, monthly, and it is an average across whoever was surveyed in that geography that month.

Which matters more than it sounds, because PSA runs three separate hog surveys and they are three different populations, not three attempts at the same number:

Three answers for the same month, ₱2.10 apart. None of them is wrong. If you are a backyard raiser, the household table is your population and it reaches back sixteen years, which the all-farms table does not. Its first row is January 2010 at ₱90.26, so backyard farmgate has not quite doubled in sixteen years while feed has done considerably more than that.

One thing the household series shows that surprises people: in June 2026 the backyard population was paid slightly more than the commercial one. Backyard raisers are not automatically the ones getting squeezed on price. They are the ones getting squeezed on cost. We keep the two populations side by side on the backyard versus commercial page precisely so nobody has to average them.

The lag, and the blank month

The survey has to be collected, cleaned and published, and that takes roughly two months. Reading this in late September 2026, the newest hog month with figures is June.

July 2026 is a different thing entirely. PSA published it and published nothing in it: every one of the 110 geographies in the hog series carries a July row with no value. Our archive stores that as null.

So if a page quotes you a PSA hog price for "last month", it is not a PSA figure. It is either a DA bulletin, a trade quote or somebody's guess. All three are legitimate things to publish. None of them is the national statistical series, and the difference is worth pesos.

Liveweight, not dressed

The PSA hog series is liveweight at the farm gate. The whole animal, on the scale, alive.

Retail pork prices are dressed and cut. Dressing yield on a Philippine hog runs roughly 65% to 75% depending on breed, condition and how the carcass is trimmed, so there is no clean multiplier between the two and anybody who gives you one is rounding. If you want the full version of that argument, we wrote it up separately: liveweight versus dressed weight.

The practical use is defensive. When a buyer quotes you a number, ask which weight it is on before you say yes. A quote that sounds generous on dressed weight is a poor one on liveweight and the conversation is much harder to reopen afterwards.

Trap one: the province that appears twice

Republic Act 12000 created the Negros Island Region, moving Negros Occidental, Negros Oriental and Siquijor out of Regions VI and VII, and moved Sulu out of BARMM. PSA's geolocation dimension still carries both copies of each. The old rows have not been deleted. They have been left in place and they are completely empty from January 2025 onward.

Read the Region VII copy of Siquijor and you get a blank column and conclude PSA does not publish Siquijor. Read the Negros Island Region copy and you get 279.24, the highest provincial figure in the country. Same province, same month, two rows, and only one of them has anything in it.

Sulu is the honest exception: it carries two rows and both are empty, so this series has no Sulu figure at all. That is a fact about PSA's survey coverage, not about Sulu.

The undivided Maguindanao row is the same kind of ghost, superseded when the province split into Maguindanao del Norte and del Sur, both of which do publish.

This is why our count is 86 provinces with a June figure out of 92 rows. Six of the 92 are administrative leftovers, not missing data.

Trap two: the number that is a quote, not a survey

Here is a diagnostic nobody else publishes, and it takes ten seconds.

Pull six months of your province and count how many different values appear. A real market surveyed monthly produces six different numbers, or at worst five. A province that produces one or two is not being surveyed month to month. It is carrying an administrative quote forward until somebody updates it.

Sixty-five of the 86 provinces pass this test with six distinct values. Five fail it badly. And two of those five sit inside the country's top eighteen by price, which means a straight ranking of provinces puts a carried quote above real markets and nobody says so.

This does not make the flat figures useless. It makes them anchors rather than readings. If you farm in Ifugao, 240.00 is a real number somebody at some point stated; it is just not evidence of what happened last month.

Trap three: do not derive one level from another

This one caught us, so we are passing it on.

PSA published seventeen regional hog figures for June 2026. Take their unweighted mean and you get ₱175.14. PSA's own published national figure for that month is ₱171.76. Those do not match, and they are not supposed to: the national figure is not a plain average of the regional ones you can see.

The same applies going down a level. Central Luzon's published regional figure is ₱181.05/kg for June 2026. The simple mean of its seven provinces is ₱177.72. Different again.

So the rule: quote the level you actually want, from the row that publishes it. Do not compute a regional figure from provinces, do not compute a national figure from regions, and be suspicious of anybody who does. If you need to explain a regional figure in terms of its provinces, say you are inferring weights rather than reporting a statistic. We do that occasionally and we label it every time.

Trap four: the official number and the trade number are not the same number

This is where most price arguments in the industry actually live.

SINAG told BusinessWorld in May 2026 that live hogs had averaged ₱180 to ₱190 a kilo over the previous seven months. PSA's own monthly series for that identical window, November 2025 through May 2026, runs 177.78, 187.06, 172.56, 176.31, 179.23, 174.00 and 171.56. Those seven months average ₱176.93. One of the seven reached ₱180 at all.

Both are honestly reported. They are not measuring the same thing: a survey of what producers received against an industry body's read of what the trade was paying. The gap is worth ₱3 to ₱13 a kilo to somebody deciding whether an offer is fair, which on a 95 kg pig is ₱285 to ₱1,235.

The same discipline applies to a biyahero's quote, a Facebook price-monitoring post and a DA regional bulletin. All are useful. None is PSA. Label them and you will stop arguing in circles.

Trap five: the survey is not your municipality

A provincial figure is an average across a province. Yours is a single farm in a single town with a particular buyer.

In August 2026 hog raisers in South Cotabato told the Inquirer they were being paid ₱110 to ₱120 a kilo. PSA's South Cotabato row for June is 137.25. Both can be true: the official row is two months older and describes a whole province, the trade read is current and describes a handful of farms. In a falling market the official figure is the ceiling of what you might expect, not the floor. We walked through what that means for Mindanao separately.

Use the published row to know whether an offer is roughly in the right neighbourhood. Use two live quotes to know what this week's actual price is. The two jobs are different and neither substitutes for the other. Our price page keeps both sets side by side and updates them monthly.

What we publish, and why we publish our gaps

Everything above is in our archive, downloadable, with provenance on every row: the series id, the date the row refers to, the value or an explicit NA, the geolocation code, the provenance class, the source URL, and the date a human actually opened that URL. The column order is a published interface with a unit test asserting it, so you can script against it and it will not move under you.

We also publish what we do not have. The gaps page lists every known hole and every retracted figure with the reason, including the ones where we got something wrong and pulled it. A dataset with no stated gaps is a dataset whose gaps you have not been told about.

Kung dili nimo makita ang numero sa imong probinsya, ayaw pag-imbento. If you cannot find the figure for your province, do not invent one. Write the honest range and say it varies by area. That is the whole discipline, and it is the only reason any of this is worth anything.

One habit worth building

Open the table once a month, on the same day, and write down two numbers: your province's newest published figure and the best live quote you can get from a buyer. Keep them in the same notebook you keep your feed receipts in.

After six months you will have something no article can give you, which is the gap between what the survey says about your province and what your particular buyer actually pays you. That gap is your negotiating position, and it is the only price number on this whole page that is genuinely yours.