The Palm Oil Question

We use palm shortening. Here’s the health evidence, the deforestation evidence, and the honest reasoning behind choosing a supplier anyway

Palm oil is the ingredient most likely to make someone put a product back on the shelf, and I understand exactly why. So rather than bury it in an ingredient list, I want to lay out what we found when we actually went looking — including the parts that don’t flatter us.

We’re consolidating our shortening onto a single palm-based supplier, Olera. This is the reasoning.

First, palm oil is not coconut oil

The phrase “tropical oils” does a lot of damage by lumping together fats that are chemically unrelated.

Palm oil is roughly 49% saturated, 37% monounsaturated and 9% polyunsaturated. Its single largest fatty acid is very often oleic acid — the same monounsaturated fat that dominates olive oil — at 36–44%, with palmitic acid at 39–47%.

Now compare. Coconut oil and palm kernel oil are both around 82% saturated. Butterfat, on the same basis, is about 63%. Beef tallow is roughly 50%.

So palm oil is about half saturated — meaningfully less saturated than butter or ghee, dramatically less than coconut, and far more than canola at 7%. Palm and palm kernel oil come from the same fruit and are not remotely the same fat; the kernel is a lauric fat, the flesh is not.

What the health evidence actually shows

Not “palm oil is fine.” Something more specific.

Pooled controlled-trial data shows that when palm oil replaces oils rich in monounsaturated fat, LDL cholesterol rises by about 9 mg/dL. Replacing polyunsaturated oils, about 3 mg/dL. Both real, both modest, both in the direction you’d expect.

But when palm oil replaces other saturated fats, LDL goes down slightly — about 2 mg/dL. And in the head-to-head that matters most for a bakery, coconut oil raises LDL by roughly 20 mg/dL more than palm oil does.

That’s the honest shape of it. Against liquid unsaturated oils, palm looks worse. Against the solid fats it actually competes with in a bakery — butter, coconut, tallow — it looks the same or slightly better.

I’ll give you the counterweight too, because I’d rather you hear it from us. The American Heart Association does specifically name palm oil, palm kernel oil and coconut oil among fats to limit, and the 2025 Dietary Guidelines Advisory Committee graded the evidence as moderate that replacing palm oil with unsaturated vegetable oils lowers LDL. Meanwhile the World Health Organization’s guidance never names palm oil at all — it addresses saturated fat as a category, recommending it stay under 10% of calories, while noting these recommendations “do not preclude consumption of particular foods.”

Nobody credible claims palm oil is uniquely dangerous. The guidance is about saturated fat generally, and cake is not health food. We’ve never pretended otherwise.

One thing I won’t skip. There is a genuinely palm-specific issue: 3-MCPD and glycidyl esters, contaminants formed during high-temperature refining. Palm is the worst affected because of its natural composition, and European regulators flagged it. It’s worth knowing the trajectory, though — levels of glycidyl esters in palm oils halved between 2010 and 2015, the EU set legal maximum limits in 2018 and 2021, an international code of practice for reducing them was adopted in 2019, and refining changes have been shown to cut them by up to 90%. When EFSA revised its safe intake level in 2018 it concluded intake was safe for most consumers, with the residual concern concentrated in infants fed exclusively on formula. That’s the accurate version: a real issue, actively managed, and not about adults eating cake.

Now the harder part

Here’s where I’d rather not be defensive, because the biodiversity harm is real and documented.

The IUCN’s assessment identifies 405 species for which oil palm is cited as a threat — 193 of them Critically Endangered, Endangered or Vulnerable. Areas suitable for expansion hold 54% of the world’s threatened mammals and 64% of threatened birds. In Malaysian Borneo, oil palm has driven as much as half of all deforestation. None of that is in dispute, and no amount of context makes it acceptable.

And yet the same IUCN report reaches a conclusion most people find surprising: refusing palm oil would likely displace rather than prevent the damage.

The reason is yield. Producing a tonne of oil takes about 0.33 hectares with oil palm. Sunflower needs roughly four times that land. Soybean about seven times. Coconut about eleven. Olive about thirteen. Palm supplies roughly a third of the world’s vegetable oil — closer to 37% counting palm kernel — from under a tenth of the land given to oil crops.

The fair objection is soybean: about 80% of a soybean becomes protein meal, not oil, so charging all its land to oil overstates its inefficiency. That’s true, and worth stating. It also doesn’t change the arithmetic if what you need is oil.

Modelling published in 2024 put numbers on the substitution: replacing palm with soybean, rapeseed and sunflower oil would put an estimated 28 to 52 million additional hectares of forest at risk, while leaving global emissions roughly unchanged. Moving instead to deforestation-free palm would cut emissions from that supply chain by about 92%.

So “palm oil free” isn’t a solution. It’s a relabelling that moves the deforestation somewhere with a weaker campaign against it.

What’s actually happening to the forest

The trend is better than most people think, and more fragile than the industry admits.

Industrial oil-palm-driven deforestation in Indonesia peaked around 227,000 hectares in 2012. By 2022 it had fallen to about 22,000 — roughly a 90% decline. Then it rose again: about 31,000 hectares in both 2024 and 2025.

Two things stop that from being a victory lap. First, peer-reviewed analysis found that a 1% fall in palm oil prices was associated with a 0.68% decrease in forest loss — meaning a meaningful share of the improvement was economics, not governance, and prices can rise again. Second, the frontier is moving: deforestation in Papua more than doubled between 2024 and 2025.

Improving, reversible, and not finished. That’s the accurate summary.

Does certification actually work?

Partly. Less than the logo implies.

The landmark study found RSPO certification reduced deforestation by about a third — from 9.8% to 6.6% per year in certified plantations. That’s real.

But the same study found certification is applied disproportionately to land that was already cleared: certified plantations held 12% forest cover at baseline against 36% for uncertified ones. It found no significant effect on fire, and none on peatland clearing. A separate study in Kalimantan found certified plantations no better at protecting orangutans, with no clear improvement in neighbouring villagers’ wealth or health access.

And the most uncomfortable finding is the newest. A 2026 analysis of corporate zero-deforestation pledges in Indonesian palm oil found companies largely kept them — under 1% annual deforestation in covered concessions — but that clearing fell just as fast in areas not covered. High compliance, no additionality. The pledges couldn’t be credited with the improvement.

RSPO covers about 20% of global supply, a share that has barely moved in eight years. Certification is a floor, not a fix. Anyone selling it to you as absolution is overselling.

So why Olera

Given all that, here is our actual reasoning, without varnish.

Function. Palm is naturally semi-solid with the beta-prime crystal behaviour bakery fats need, achieved without hydrogenation. The alternatives each cost something: coconut is functionally awkward and worse for LDL; butter carries water and is the most expensive option; fully hydrogenated and interesterified fats work well and the evidence says they’re metabolically neutral, but they’re a harder ingredient to explain honestly on a label. Olera’s shortening is non-hydrogenated and contains no partially hydrogenated oils — the compositional claim we can verify directly from specification.

Direction. If boycotting palm displaces the harm, the useful lever is demand for the better-governed end of the supply. Olera’s parent, Global Agri-trade Corporation, is a member of the Roundtable on Sustainable Palm Oil — membership number 2-0156-10-000-00, in the Ordinary Palm Oil Processors and Traders category. That we confirmed in RSPO’s own records.

And here’s what we could not confirm. Olera states on its website that it supplies at the RSPO Segregated level — the stricter supply-chain model, where certified oil is kept physically separate rather than merely offset. We could not independently verify that for the specific product we buy. No certificate number, no named certifying body, and no expiry date is published anywhere we could find. RSPO’s own historical filing for the parent company shows a mixed portfolio across supply-chain models rather than fully segregated volume.

So we’re asking Olera for their current supply-chain certificate number, the accredited body that issued it, and written confirmation of the supply-chain model for the specific product we buy. We’ll publish what comes back — including if what comes back is unsatisfying.

That felt like the only defensible way to write this. A brand that tells you to read labels carefully doesn’t get to reprint a supplier’s marketing bullet as though it were a fact.

Where these numbers come from

• Fatty acid composition of palm, palm kernel, coconut, canola, sunflower: Codex Alimentarius (FAO/WHO), Standard for Named Vegetable Oils CXS 210-1999 — fao.org
• Saturated/monounsaturated/polyunsaturated totals for palm (49/37/9), coconut and palm kernel (82% saturated), dairy fat (63%), tallow, canola: Sacks FM et al., “Dietary Fats and Cardiovascular Disease: A Presidential Advisory From the American Heart Association,” Circulation 136:e1–e23 (2017) — ahajournals.org
• LDL effects of palm oil versus monounsaturated (+9.18 mg/dL), polyunsaturated (+3.43 mg/dL) and other saturated fats (−2.32 mg/dL): Unhapipatpong C et al., “Tropical Oil Consumption and Cardiovascular Disease: An Umbrella Review,” Nutrients 13(5):1549 (2021) — mdpi.com
• Coconut oil raising LDL relative to palm oil: Neelakantan N et al., Circulation 141:803–814 (2020)
• WHO guidance on saturated fat (10% of energy; no naming of individual oils): World Health Organization, Saturated fatty acid and trans-fatty acid intake for adults and children: WHO guideline (2023) — who.int
• AHA naming palm, palm kernel and coconut oils among fats to limit: American Heart Association — heart.org
• Moderate-grade evidence on replacing palm oil with unsaturated oils: 2025 Dietary Guidelines Advisory Committee Report, Part D, Chapter 4 — dietaryguidelines.gov
• Glycidyl esters halving 2010–2015; revised 3-MCPD tolerable intake and conclusion on consumer safety: European Food Safety Authority (2016 and 2018) — efsa.europa.eu
• EU maximum levels for glycidyl esters and 3-MCPD: Commission Regulations (EU) 2018/290 and (EU) 2020/1322
• International code of practice for reducing 3-MCPD and glycidyl esters: FAO/WHO Codex Alimentarius, CXC 79-2019
• 405 species threatened, 193 Critically Endangered/Endangered/Vulnerable; palm supplying ~35% of vegetable oil on under 10% of oil-crop land; conclusion that substitution displaces rather than halts biodiversity loss: IUCN Oil Palm Task Force, Oil palm and biodiversity: A situation analysis (2018) — portals.iucn.org
• Land required per tonne of oil by crop: Our World in Data, computed from FAOSTAT (2023) — ourworldindata.org
• Substitution modelling (28.2–51.9 million additional hectares at risk; 92% emissions reduction from deforestation-free palm): Chiriacò MV, Galli N, Santini M, Rulli MC, Science of the Total Environment 914:169486 (2024)
• Indonesian palm-driven deforestation 2012 peak, 2022 trough and 2024–25 figures; Papua doubling: TheTreeMap / Nusantara Atlas — nusantara-atlas.org
• Price elasticity of forest loss: Gaveau DLA et al., PLOS ONE 17(3):e0266178 (2022) — journals.plos.org
• RSPO certification reducing deforestation by 33%; baseline forest cover of certified versus uncertified plantations; no effect on fire or peat: Carlson KM et al., PNAS 115(1) (2018) — news.essic.umd.edu
• Certification showing no measurable benefit for orangutans or village wellbeing: Morgans CL et al., Environmental Research Letters 13 (2018)
• Zero-deforestation commitments showing high compliance but no additionality: Stigler M et al., PNAS (2026), DOI 10.1073/pnas.2511503123
• Evidence that interesterified fats do not affect fasting serum lipids: Mensink RP, van Rooijen MA, Nutrients 12(3):615 (2020)
• Global Agri-trade Corporation RSPO membership number and category: RSPO member records and the company’s Annual Communication of Progress filing hosted by RSPO
• Olera’s own statement regarding RSPO Segregated supply: olera.com/sustainability and olera.com/palm-oil-palm-kernel-oil

Every figure above comes from government or intergovernmental bodies, peer-reviewed research, or the IUCN, except where we have explicitly identified a claim as our supplier’s own statement rather than an independently verified fact.