Wild bees vs. honeybees

Start with the species

When most readers hear “bees,” they picture the western honeybee, Apis mellifera. That is one species. There are roughly 20,000 other bee species in the world — bumblebees, solitary mining bees, mason bees, leafcutter bees, sweat bees, carpenter bees, stingless bees, and many more — collectively called wild bees. Apis mellifera is wild in parts of its range too, but the population most people interact with is managed: it lives in human-made hives, is moved around for honey production and crop pollination, and is propagated by beekeepers. The IPBES Assessment Report on Pollinators, Pollination and Food Production (2016)^[1] is the authoritative consensus document on the global pollinator picture; its summary for policymakers is the most-cited single source on the role of pollinators in food production and ecosystems.

Almost every public number you have ever read about “bees” refers to one of these two populations, not both. They behave differently, they are measured differently, they face different threats, and they need different policy. The site you are reading exists in part to keep them straight.

What managed honeybees do

Managed Apis mellifera are an agricultural livestock animal. They are kept in stacked-frame hives, fed sugar syrup or pollen substitute when nectar is short, treated for Varroa destructor mites with miticide strips or vapours, requeened on a schedule, and transported by truck to crop-pollination contracts (in the US, almonds in California are the canonical example; in Europe and Asia, the mix differs). Their population at the country level is reported annually to FAOSTAT as element 5111, “Live animals — Bees (beehives)”, and the FAOSTAT total has been rising for decades. As of the most recent published year, FAOSTAT reports tens of millions of beehives globally — the long-run global trend is up^[2].

That fact — managed honeybee colonies are globally up over the long run — is regularly cited by people arguing that “the bee crisis is overblown.” The data they cite is correct. The inference they draw from it is wrong, because it does not apply to wild bees, and because country-level totals hide colony-loss problems within them.

What “colony loss” surveys actually count

The most-cited bee statistic in the United States is the annual colony-loss survey from the Bee Informed Partnership (BIP)^[3], conducted with the Apiary Inspectors of America. It asks US beekeepers how many of their managed colonies died over the past year. The headline number has hovered around 30–50% annual losses for most of the past 15 years. That number is real and it is bad: it means most US commercial beekeepers replace roughly a third to a half of their hives every year just to keep their headcount stable.

Crucially, that survey is about managed colonies, not species. A 40% loss followed by a 45% replacement (via splits and new queens, the standard practice) shows up in FAOSTAT as a stable or growing hive count. So both statements are simultaneously true: US managed-honeybee colony loss rates are alarmingly high, and the global FAOSTAT beehive count is at an all-time high. They are measuring different things. Annual loss is a measure of stress on beekeepers; FAOSTAT is a measure of beekeeper persistence in the face of that stress.

The corresponding survey work has been replicated by the international COLOSS network across more than 30 countries, with broadly similar findings: managed colonies are under heavy annual stress in most of the well-monitored beekeeping world, and the loss rate is variable but persistently high^[4].

What is happening to wild bees

The wild-bee picture is different again, and the data is patchier. The most-cited authoritative source is the European Red List of Bees, an IUCN regional assessment of 1,965 European wild bee species published in 2014 by Nieto et al.^[5]. About 9% of all assessed European bee species were classified as threatened (Critically Endangered, Endangered, or Vulnerable), and another ~5% as Near Threatened. The corresponding figure for bumblebees alone was higher: about 24% of European bumblebees were classified as threatened. Crucially, roughly 57% of the assessed species were Data Deficient — the assessors had too little data to evaluate them at all — so the true threatened share is likely higher than 9%.

In the United States, the rusty patched bumblebee (Bombus affinis) was federally listed as endangered under the Endangered Species Act in 2017 — the first bee in the contiguous US to be so listed^[6]. The eastern range of several other Bombus species (the American bumblebee, the western bumblebee) is contracting on a scale measurable in degrees of latitude over a few decades^[7].

Globally, the IPBES 2016 assessment was unable to give a single headline number for wild-bee decline because the data simply does not exist for most of the world — large parts of tropical Africa, tropical Asia, the Neotropics, and Oceania have so little bee survey data that any global figure would be guessing. The IPBES summary instead reports that where good data exists, wild pollinator declines are documented, and that the threats (habitat loss, pesticide exposure, pathogens spilling over from managed bees, and climate-driven range shifts) are well understood^[1].

Why the two stories diverge

Managed honeybees survive in spite of all the same threats because beekeepers actively replace them. A wild bee species has no beekeeper. When a wild bumblebee population in upstate New York or Oxfordshire collapses, nobody splits a hive to replace it. When pesticide-treated forage suppresses larval survival across a species’ range, there is no miticide-strip equivalent to deploy.

This is also why the actions that protect one population do not necessarily protect the other:

• Importing more managed honeybees into an area to “help the bees” can actually harm local wild bees through competition for forage and pathogen spillover. This is documented in protected-area contexts in several biomes^[8].
• Banning a specific pesticide (e.g. the EU’s 2018 outdoor-use restriction on three neonicotinoids^[9]) benefits both managed and wild bees, but the wild-bee benefit is harder to measure because wild-bee monitoring is patchier.
• Planting native flowering forage benefits wild bees disproportionately, because wild species tend to be more specialised on native plants than honeybees, which forage opportunistically on almost anything.
• Protecting and connecting habitat at scale — hedgerows, meadow strips, urban park corridors — benefits wild bees almost exclusively, because managed honeybees are by definition not habitat-limited (they live in human-supplied housing).

What this means for what you read

If a public-interest story about bees does not specify which bees, it is almost certainly conflating the two. Some heuristics:

• If the source is FAOSTAT, USDA, NASS, or a national agriculture department, the story is about managed Apis mellifera. Treat it as livestock data.
• If the source is a colony-loss survey (Bee Informed Partnership, COLOSS), the story is about beekeeper-reported loss of managed colonies. It says nothing about wild species.
• If the source is an IUCN Red List assessment or a national list (e.g. Species at Risk Act listings in Canada, ESA listings in the US, the European Red List), the story is about wild species. Each species has its own range, threats, and status.
• If the source is GBIF (or iNaturalist data flowing into GBIF), the story is about wild-bee observation records, biased heavily toward where citizen scientists and entomologists are active. Sampling effort drives the numbers as much as biology does.
• If the source is IPBES, the FAO Pollinators Initiative, or a peer-reviewed synthesis paper, the story is the carefully-constructed both-of-them picture, and is worth reading slowly.

What to actually do

The actions on the country map are tagged to which population they help most. Plant native pollinator forage, submit one bee photo to GBIF via iNaturalist, and defend existing pollinator habitat primarily benefit wild bees. Support the Bee Informed Partnership (US colony-loss research) and find and support a local beekeeping association primarily benefit managed honeybees. Cut pesticide pressure and tracking the EFSA and US EPA pesticide reviews benefit both. Read primary sources benefits you.

The single most-leveraged change a casual reader can make is to stop treating these as the same problem. Once you separate them, the science becomes much easier to follow, the news much easier to evaluate, and the action much easier to choose.

Sources

1. IPBES (2016). Summary for policymakers of the assessment report of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services on pollinators, pollination and food production. https://www.ipbes.net/assessment-reports/pollinators
2. FAOSTAT — Live animals (beehives), element 5111. https://www.fao.org/faostat/en/#data/QCL
3. Bee Informed Partnership — annual US colony-loss surveys. https://beeinformed.org/
4. COLOSS — international honey bee research association, multi-country annual surveys. https://coloss.org/
5. Nieto, A., Roberts, S. P. M., Kemp, J., Rasmont, P., Kuhlmann, M., et al. (2014). European Red List of Bees. Luxembourg: Publication Office of the European Union. https://www.iucnredlist.org/resources/european-red-list-of-bees
6. US Fish & Wildlife Service — Rusty Patched Bumble Bee (Bombus affinis) species page. https://www.fws.gov/species/rusty-patched-bumble-bee-bombus-affinis
7. Cameron, S. A. & Sadd, B. M. (2020). Global trends in bumble bee health. Annual Review of Entomology 65: 209–232. https://doi.org/10.1146/annurev-ento-011118-111847
8. Geldmann, J. & González-Varo, J. P. (2018). Conserving honey bees does not help wildlife. Science 359(6374): 392–393. (Representative; many peer-reviewed reviews now exist on managed-honeybee competition with wild bees in protected areas.) https://doi.org/10.1126/science.aar2269
9. European Food Safety Authority — Neonicotinoids and bees, 2018 conclusions. https://www.efsa.europa.eu/en/press/news/180228

This primer is published under CC BY-SA 4.0. Copy, translate, and republish freely; please keep the citations intact.