Is the Backyard Pig Raiser Disappearing? What PSA's Numbers Actually Show

There is a stock paragraph in almost everything written about Philippine hog farming since 2021. It says African swine fever wiped out the backyard raiser, that the survivors are commercial farms, and that the smallholder is on the way out.

PSA counts every pig in this country four times a year and publishes the result by farm classification. It says the opposite.

On 1 July 2026 the smallhold classification held 7,080,312 head, which is 79.3% of the national herd and the highest share in a series running back to January 2010. The commercial classification held 1,698,801, down 24% on the year. If somebody is disappearing here, it is not the person with eight pigs behind the house.

First, the trap

Two PSA tables, and if you pick the wrong pair of rows you will publish a collapse that did not happen.

They are not two versions of one thing. PSA replaced a classification based on farm character with one based on farm size, and it created a third bucket in the process. A farm that counted as Backyard in 2020 is not automatically Smallhold now. We keep them in separate series for exactly this reason: legacy backyard, legacy commercial, smallhold, semi-commercial and commercial, each with the table path it came from on every row.

The standard warning is that splicing them draws a collapse in backyard farming that is really just the definition moving. That warning is correct and this site has printed it repeatedly.

It is also, as far as I can find, never quantified. So let us quantify it.

How big the definitional break actually is

Both tables publish the eight January mornings from 2013 to 2020. Same country, same mornings, two classifications. Put them side by side.

Two things fall out of that table and both are worth having.

The national totals are identical to the head, every year. PSA restated the same survey under a new classification rather than running a new survey. That means the comparison above is a clean read on the reclassification and nothing else.

The reclassification is small. The largest gap in the eight years is 2.06% of the backyard count, in 2017. The smallest is 0.32%, in 2019, which is 26,239 head out of 12.7 million. And in 2020 it goes the other way, with the new scheme counting 128,586 more smallhold head than the old scheme counted backyard.

What actually happened to the herd

Now the current series on its own, quarterly, no splicing, from its own first morning to its newest.

The national herd is down about 30% from 12,699,991 head on 1 July 2019 to 8,928,329 on 1 July 2026. It also grew in the most recent quarter, up 2.6% from the 8,702,715 counted on 1 April 2026. Everybody prints the first of those two facts. Almost nobody prints the second.

Underneath the total, the two populations went in opposite directions.

Commercial. Peaked at 4,436,104 head on 1 October 2018, three months before ASF was first confirmed here. It is now 1,698,801, down 62% from that peak and down 24% in the last twelve months alone. It is still falling, quarter after quarter, six years after the outbreak began.

Smallhold. Bottomed at 6,102,858 on 1 January 2025 and has risen in five of the six quarters since, to 7,080,312. Up 9.7% year on year. It is down about 23% from its January 2010 level, which over sixteen years is barely a trend at all.

Sus. That is not the story anybody has been telling.

Why the commercial farm is the one bleeding

I am not going to pretend the archive settles this, because it does not. But the pieces that are in it point the same way, and they are worth laying out.

A commercial farm carries fixed costs a backyard pen does not. PIDS costed both in 2022 and found operating costs at 68% of gross returns for a commercial operation and 89% for a backyard one, which sounds like the commercial farm has the cushion. The catch is what happens when the price falls below cost: a backyard raiser stops buying weanlings for a cycle and eats the loss in unpaid labour, and the pen sits empty until things improve. A commercial farm with a loan, a payroll and depreciating housing cannot idle. It either runs at a loss or it liquidates the sow herd, and a liquidated sow herd takes two years to rebuild.

Biosecurity cuts the same way. ASF is far more lethal to a dense operation than to three pigs in a backyard, and the restocking calculus after a wipe-out is completely different at 1,400 head than at 40.

And then there is the price. The national farmgate average has been under pressure all year, and a farm whose whole business is selling hogs at that price has nowhere to hide. A household selling four pigs a year has a job.

The thing this data cannot tell you

Here is the honest limit, and it goes to the heart of the question in the title.

PSA counts pigs, not farmers. The inventory table reports head by farm classification. It does not report how many farms there are, and neither does anything else this archive holds. So a rising smallhold head count is compatible with two completely different worlds: more households keeping pigs, or the same number keeping more, or fewer households keeping many more each.

That last one is not far-fetched. A smallhold classification with 7.08 million head could be five hundred thousand households with fourteen pigs each or two million households with three and a half. Those are radically different industries and the published series cannot separate them.

The second limit is the informal chain. PSA's registered-slaughter series counted 9,412,836 hogs through registered slaughterhouses in 2025, against national production of 1,658,283 tonnes liveweight. Divide the tonnage by a plausible market weight and you get somewhere between 15.8 and 17.5 million head. So roughly 54% to 60% of the country's hog production passes a registered abattoir and the rest goes through backyard slaughter, community butchering and unregistered facilities. That figure is ours and the market-weight assumption is doing real work in it, so treat it as a band rather than a number. Either way, a very large slice of the smallhold sector is invisible to every series that tracks the formal chain.

The four choices

So what does a person with a small pen actually do. There are four, and three of them are worse than they look.

Restock

The case for it is that the herd is still 30% under its pre-ASF level, the smallhold segment is the one growing, and land and labour you already own are sunk.

The case against it is that everybody else has had the same idea. That 9.7% year-on-year rise in the smallhold herd is your neighbours restocking, and it showed up in a 6% production rise and a 17% price fall in the same half-year. The correct question is not whether the herd has room. It is whether your cost per kilo clears your province's actual farmgate figure. If it does not, restocking is buying a loss.

Scale

Going from ten pigs to fifty spreads your fixed costs and gets you a better feed price and a better buyer. It also multiplies whatever your per-kilo margin currently is, in whichever direction it points.

Look at that commercial column again before you do it. The segment with the best cost per kilo in the country, at PIDS's ₱78.68 against ₱103.78 for backyard, is the segment that has lost 62% of its herd since 2018. Scale did not save them. Scale is a lever on a margin, and if the margin is negative the lever works fine.

If you are going to scale, scale the input side first. A bigger volume buys a better feed price, and feed is the line with the most give in it: the same grower recipe cost 21% more to mix in July 2026 than a year earlier.

Specialise

This is the only one of the four that changes the price you are paid rather than the cost you pay.

Native pigs for lechon de leche, a breeder operation selling weanlings into a restocking market, direct selling into your own barangay at retail rather than through a viajero, contract growing where somebody else carries the price risk. Each of these takes you off the national farmgate line, which is the line with a ceiling on it that imports put there.

They are all harder than they sound and most of them need a permit, a market you have already tested, and a year of somebody else's money. But the arithmetic in our price chain piece is unambiguous about where the money in this industry currently sits, and it is not at the farm gate.

Exit

Nobody else will say this, so I will. If your break-even is well above your province's published figure, you have no route to a better price, and the pen is your only asset, then selling the stock and stopping is a legitimate answer and not a failure. The commercial sector has made that decision 2.7 million head worth of times since 2018 and nobody calls those farms quitters.

What is not legitimate is drifting. Running a pen at a loss for three more cycles because stopping feels like giving up is how people lose the land as well as the pigs. Whether a piggery is a good investment at all is a question worth answering honestly before the next batch of weanlings, not after.

Dili ang backyard ang nahanaw. Ang dagkong commercial farm ang nawad-an ug 62 porsyento sukad 2018. Apan ang kadaghanan sa mga sulat bahin niini, baliktad ang pagbasa.

What to actually watch

One number, once a quarter, and it is free. PSA publishes the swine inventory on the first morning of January, April, July and October. Open the smallhold row and the commercial row and write both down.

If the smallhold line keeps rising while the commercial line keeps falling, the industry is being handed to smallholders by default, and the constraint on your business is the price rather than the competition. If the commercial line turns and starts climbing, somebody with capital has decided the price will recover, and they will be at your buyer's gate before you are.

Either way you will know it two months before it reaches an article, which is the whole reason we publish the raw series rather than a chart of it.