Valuing over-winter colony losses for New Zealand’s commercial beekeepers

ABSTRACT

The number of managed honeybee colonies in New Zealand has increased dramatically, due largely to increasing high-value mānuka honey exports. Pollination also represents an important part of the apicultural sector, providing critical services to the horticultural and arable crop industries. A survey of winter colony losses has been conducted annually since 2015. Using data from the 2021 wave of the survey, we derive a model to estimate the financial cost of winter colony loss to New Zealand commercial beekeepers in that year. We estimate that winter colony losses conservatively cost commercial beekeepers $24,181,835 in 2021, which amounts to $38.04 per managed colony. This equates to 9.2% of the value of honey yields for commercial beekeepers who produce honey and between 10.6% and 60.5% of the value of contracts for beekeepers who provide pollination services. These substantial costs pose a risk to the profitability of beekeeping.

KEYWORDS:

• Apiculture
• varroa
• honey
• pollination

Introduction

At a meeting of the National Beekeepers’ Association in 1950, the Minister of Agriculture and Marketing (and future Prime Minister) Keith Holyoake emphasised the importance of New Zealand’s apiculture industry, noting that the country’s 6487 beekeepers owned 174,386 honeybee (Apis mellifera) colonies and produced 4517 tons of honey (Gisborne Herald, 1950). Seventy-two years later, 9124 beekeepers (an increase of 41%) manage 725,210 honeybee colonies (an increase of 316%), making New Zealand’s apiculture industry the world’s most commercialised.¹ New Zealand beekeepers produced approximately 20,500 tonnes of honey in the 2020/2021 season, and earned $482 million in export revenue in the year ended 30 June 2021 (Ministry for Primary Industries, 2021), largely driven by demand for mānuka honey with its documented antimicrobial properties (Girma, Seo, & She, 2019; Johnston, McBride, Dahiya, Owusu-Apenten, & Nigam, 2018). In addition, managed honeybees provide critical pollination services to the horticulture and arable crop industries – services that were valued at approximately $5 billion in 2014 (New Zealand Parliament, 2014).

The persistent threat of diseases and pests – especially the parasitic mite Varroa destructor – has compelled many countries to undertake periodic surveys of over-winter honeybee colony losses. Losses vary dramatically by region and year; for example, a study of 2019/2020 over-winter loss rates in 37 mostly European countries showed that loss rates ranged from 7.4% in Norway to 36.5% in Spain (Gray et al. 2022). Spain’s loss rates jumped 49.0% relative to the previous year (Gray et al. 2020).

The New Zealand Colony Loss Survey (NZCLS) has been conducted annually since 2015 (Brown, Newstrom-Lloyd, Foster, Badger, & McLean, 2018). While over-winter loss rates in New Zealand are relatively low by international comparison, they have increased monotonically, from 8.4% in winter 2015–13.6% in winter 2021 (Stahlmann-Brown, Hall, Pragert, & Robertson, 2022). If 2015 loss rates had persisted, New Zealand would have lost approximately 42,000 fewer colonies over winter in 2021. Varroa alone was the attributed cause of death of an estimated 5.3% of all living honeybee colonies in winter 2021 (Stahlmann-Brown & Robertson, 2022).

While the economic impacts of any losses are potentially significant, the complexity of beekeeping makes the value of colony losses difficult to estimate. We know of only one study that has attempted this, valuing colonies lost over winter 2016/2017 at € 32,031,305 in Austria, € 21,400,401 in Czechia, and € 3,038,741 in Macedonia (Popovska Stojanov et al. 2021). However, these estimates exclude lost pollination fees, labour required to replace lost or weak colonies, and lower returns stemming from changes to management practices.

In this paper, we develop a model to estimate the value of over-winter colony losses to commercial beekeepers in New Zealand, accounting for both lost honey crop and affected pollination contracts. We focus on commercial operators because of their significant contributions to the New Zealand economy. We have not considered the cost of colony losses to non-commercial beekeepers, but we recognise that the value of these losses are likely to be substantially higher on a per-hive basis because non-commercial beekeepers are less likely to benefit from efficiencies of scale (e.g. access to bulk-buying prices for varroa treatments, sufficient hive stocks to allow for colony management by splitting) and experience greater variability in colony health outcomes (e.g. small-scale beekeepers report higher colony loss rates).

Data

The NZCLS is an Internet-based survey conducted between 1 September and 15 November annually (Stahlmann-Brown et al. 2022). All New Zealand beekeepers are legally required to register their hives for biosecurity purposes under the National American Foulbrood Pest Management Plan.² In so doing, over 97% provide email addresses, and hence nearly every beekeeper in New Zealand receives a personalised invitation to complete the survey. Topics covered in the survey include over-winter loss rates, the share of colonies that survived winter but that were weak in spring, operation size, and primary use of the colony, among other topics.

In 2021, 49.1% of all registered beekeepers completed the survey; together, they managed 47.2% of all colonies in the country. These response rates are world leading³ and provide confidence in the survey results.

Accurate pricing of honey is difficult to obtain given a range of technical specifications or requirements used by honey packers and exporters when purchasing honey in addition to commercial sensitivities.⁴ We have derived average prices paid to beekeepers for monofloral mānuka,⁵ multifloral mānuka, and non-mānuka honeys based on data provided by six honey buyers on condition of anonymity. These average prices were checked against the export prices for bulk honey reported in Statistics New Zealand Export Trade Data. Honey export volume data was sourced from Statistics New Zealand. Data on honey yields per hive were sourced from the New Zealand Ministry for Primary Industries apiculture monitoring data (Ministry for Primary Industries, 2021).

The prices of brood frames, queen cells, and varroa treatment were determined from retail catalogues. All other data were derived from the expert opinion of the authors, three of whom are commercial beekeepers with between 20 and 40 years of experience each, and key sources identified in the acknowledgements.

Model and key assumptions

Given the direct and quantifiable economic benefits of the New Zealand commercial beekeeping industry and the very different economic considerations faced by hobbyist beekeepers, we focus only on commercial beekeepers with at least 250 colonies in this analysis. We explicitly assume that costs are a linear function of losses and that the levels of losses are manageable to the beekeepers due to operational scale (e.g. that the beekeeper has ready access to healthy colonies for splitting).

The 175 commercial beekeepers responding to the 2021 NZCLS managed 364,057 honeybee colonies at the beginning of winter 2021, out of a total of 687,319 registered colonies in this size class. In our simplified model, we assume that colonies are used for honey production or pollination, not both. Based on NZCLS data, 83.0% of colonies were primarily used to harvest honey and 9.5% were primarily used for pollination.⁶

Beekeepers generally have a few months between opening their hives in spring and deploying their colonies for honey production. Dead colonies may thus be replaced through splitting strong hives or making nucleus colonies from healthy hives and nursing colonies back to strength (hereafter, ‘split hives’). Split hives and weak hives still provide a honey crop, albeit with potentially lower yields than strong colonies.

For beekeepers who harvest honey, we estimate the following costs: (1) $\left[min\{\begin{array}{c}(1-\delta )2R_H+2C+Q_{sp}+nF\\ X_{sp}\end{array}\right]\times N_H^{dead}+[(1-\gamma )R_H]\times N_H^{weak}$(1) where $N_H^{dead}$ and $N_H^{weak}$ are the number of dead and weak colonies that would be used for the honey crop, $R_H$ is the value of honey produced by a strong hive, and $\delta$ and $\gamma$ are the productivity shares of split hives and weak hives relative to strong hives, respectively, with $\delta \in [0,1]$ and $\gamma \in [0,1]$. $C$ is the cost to rebuild strength in split hives, i.e. labour and additional travel for nursing. $Q_{sp}$ is the cost of a mated queen to replace the dead queen in spring. $n$ is the average number of brood frames opportunistically replaced in the dead hive and $F$ is the cost of brood frames.⁷ $X_{sp}$ is the cost of a strong hive purchased from another beekeeper in spring. We assume that beekeepers always choose the least-cost option of replacing a colony.

The value of honey produced by a strong hive, $R_H$, is $(P_{mono}\times S_{mono}+P_{multi}\times S_{multi}+P_{non}\times S_{non})\times y$ where P is the price of monofloral mānuka, multifloral mānuka, and non-mānuka as indicated by the subscript. $S$ is the quantity share of the overall yield, $y$, comprising each honey type. Lacking any data on the composition of honey produced, we note that there are three channels for honey sales in New Zealand: exports, for which official data on honey types are available; supermarket sales, which are heavily weighted toward non-mānuka but for which composition data are not available; and sales through specialty channels such as pharmacies and duty-free stores, which are similar in volume to supermarket sales but heavily weighted toward monofloral mānuka (again, composition data are not available). Assuming that the biases of supermarket and speciality sales cancel each other out, we assert that honey produced is proportional to honey exported, which, based on export data for the two-year period from 1 July 2019 to 30 June 2021 (Statistics New Zealand) is approximately 55% monofloral mānuka, 25% multifloral mānuka, and 20% non-mānuka honeys under the Ministry for Primary Industries’ manuka honey definition.⁸

The average hive yield in 2021 was 25.3 kg (Ministry for Primary Industries, 2021). Based on expert opinion, we assume that split hives produce 95% of a full honey crop ($\delta =0.95$) and that weak hives produce 80% of a full honey crop ($\gamma =0.80$).

Six buyers shared the prices at which they purchased honey (and, in some cases, volumes purchased) in 2021–2022 with the corresponding author under condition of anonymity. From this, we estimate an average bulk buy-price per kilogram (in drums) for monofloral mānuka honey of $23.90 per kilogram, an average buy-price for multifloral mānuka honey of $8.34 per kilogram, and an average buy-price for non-mānuka honey of $4.17 per kilogram.⁹ Assuming that the honey composition mirrored that of honey exports, the average price of honey harvested was $16.06 per kilogram.

We assume that splitting hives and subsequent nursing takes one hour of labour per colony at $30 per hour, on average. The price of a mated queen in spring 2021 was $60. We assume that beekeepers opportunistically replace 3 brood frames in the dead hive at $2.80 each. Accounting for the modest reduction in honey yield, the cost of splitting hives to replace dead colonies is estimated to be $169.04. The cost of a strong hive purchased from another beekeeper in spring 2021 was $250.

Beekeepers with at least 250 colonies lost 12.4% over winter 2021 (Stahlmann-Brown & Robertson, 2022). This amounts to 85,228 colonies overall. With 83.0% of colonies used primarily for honey production, we estimate that $N_H^{dead}=70,739$. These beekeepers reported that 20.4% of their surviving colonies were weak at the start of spring, therefore $N_H^{weak}=116,377$.

Honey bees also provide pollination services to the horticulture and arable sectors. In addition to rental fees and pollination dates, contracts that beekeepers enter with growers typically stipulate hive strength requirements as measured by the number of frames of bees and brood. Dead colonies that were intended for pollination may be substituted with strong hives that were otherwise intended for honey harvest.

For late-flowering crops such as rapeseed (of which approximately 3000 ha are sown annually) and carrot and onion seeds (of which approximately 300 ha are sown annually), the beekeeper with a dead colony can follow the same practice of either splitting a healthy colony or buying a healthy hive, as detailed above. We assume that weak colonies cannot be nursed back to strength in time to provide pollination services and that they will be substituted by a colony that would otherwise go into non-mānuka honey harvest.

Thus, we estimate the following costs for beekeepers providing pollination services on late-flowering crops: (2) $min\left\{\begin{array}{c}2C+Q_{sp}+nF\\ X_{sp}\end{array}\right\}\times N_{P,late}^{dead}+R_{non}\times N_{P,late}^{weak}$(2) where $N_{P,late}^{dead}$ and $N_{P,late}^{weak}$ are the number of dead and weak colonies that would be used for servicing pollination contracts for late-flowering plants and $R_{non}$ is the value of non-mānuka honey forgone as a result of the moving a strong hive from honey to pollination.

For early-flowering crops such as kiwifruit (of which 13,250 ha were planted in 2017), apples (9247 ha), avocados (4004 ha), and cherries and other summer fruit (2294 ha) (Aitken & Warrington, 2022), beekeepers lack the time required to nurse split hives back to health before pollination services are needed by growers. Hence, for both dead and weak colonies, we assume that beekeepers substitute a colony that would otherwise go into non-mānuka honey harvest. However, because a replacement colony is not required immediately, beekeepers with dead colonies can avoid the costs of requeening in spring (when prices are high) and can forego spring varroa treatments. They may also be able to purchase healthy hives later in the season, when prices are lower.

Thus, for beekeepers providing pollination services on early-flowering crops, we estimate: (3) $\left[min\left\{\begin{array}{c}2C+Q_{su}+nF\\ X_{su}\end{array}\right\}-V_{sp}+R_{non}\right]\times N_{P,early}^{dead}+R_{non}\times N_{P,early}^{weak}$(3) where $N_{P,early}^{dead}$ and $N_{P,early}^{weak}$ are the number of dead and weak colonies that would be used for servicing pollination contracts for early-flowering plants, $Q_{su}$ is the cost of a queen cell to replace the dead queen in summer or autumn, $V_{sp}$ is the cost of treating varroa in spring, and $X_{su}$ is the cost of a strong hive purchased from another beekeeper after honey flow.

Queen cells cost $10 and succeed approximately 75% of the time. Hence, the expected value of $Q_{su}$ is $13.33. Varroa treatment using the two most popular miticides, Bayvarol® and ApiVar®, costs $7.50 per brood chamber if purchased in bulk. Because almost all commercial beekeepers have two brood chambers in spring, we value $V_{sp}$ at $15. Healthy hives are steeply discounted after honey flow; hence, we value $X_{su}$ at $200. The value of non-mānuka honey produced by a strong hive, $R_{non}$, was $105.50 during the 2021/22 season.

With an estimated 85,228 colonies in total lost by commercial beekeepers in 2021, 9.5% of which were destined for pollination, we estimate that 8,097 colonies destined for use in pollination were lost. Assuming that 10% of these colonies would pollinate late-flowering crops and that 90% would pollinate early-flowering crops (consistent with the sewn areas reported above), we estimate that $N_{P,late}^{dead}=810\text{ colonies}$ and $N_{P,early}^{dead}=7,287\text{ colonies}$. With 20.4% of surviving colonies being reported as weak at the start of spring, we calculate $N_{P,late}^{weak}=1,332$ and $N_{P,early}^{weak}=11,988$.

Results and discussion

The above costs apply for New Zealand’s largest commercial operators in 2021; commercial operators with fewer hives may have less favourable cost structures, and as such, our final estimates should be considered to be the lower bound of the cost of winter colony losses to commercial beekeepers.¹⁰ Based on the above, the total cost of dead colonies to commercial beekeepers in New Zealand is approximately $13,316,835, with $11,957,835 of those costs arising from dead colonies otherwise destined for a honey crop; $103,961 in costs arising from dead colonies otherwise destined for pollinating late-flowering crops; and $1,255,040 in costs arising from dead colonies otherwise destined for pollinating early-flowering crops. The total cost of weak colonies to commercial beekeepers in New Zealand is approximately $10,864,856, with $9,459,558 of those costs arising from weak colonies destined for a honey crop; $140,530 in costs arising from weak colonies destined for pollinating late-flowering crops; and $1,264,769 in costs arising from weak colonies otherwise destined for pollinating early-flowering crops. Thus, the combined cost of winter colony losses to commercial beekeepers is thus estimated to be at least $24,181,691.

The observed rise in over-winter colony loss rates every year will have a concomitant effect on the profitability of beekeeping in New Zealand. The burden of restoring losses by splitting or nursing weak colonies back to health has the potential to exceed the net returns from honey crops or pollination services, particularly in the context of high prices for capital equipment, fuel, labour, and varroa treatment coupled with lower honey prices. This raises an important question: what winter colony loss rate can commercial beekeepers afford?

Our estimate of the financial burden of a winter colony loss to the commercial beekeeper under current colony and varroa management is $38.04 per colony – $37.54 per honey-producing colony and $42.34 per colony providing pollination services. Using our 2021 estimates of a honey buy-price of $16.06 per kg and the average honey yield per colony of 23.4 kg, we estimate that commercial beekeepers received $406.41 per colony, on average, for their honey crop in 2021. Therefore, we value the financial impact of winter colony losses at 9.2% of the value of the average honey crop. Per hive fees paid for pollination services vary between $70 and $400 (Ministry for Primary Industries, 2021). Thus, we estimate the impact of winter colony losses to be between 10.6% and 60.4% of the value of pollination contracts. These losses represent a substantial impact on the profit margin of New Zealand’s commercial beekeepers. It is important that beekeeping remains a profitable enterprise as honey generated $482 million in export revenue in 2021 (Ministry for Primary Industries, 2021) and several fruits, vegetable, and seed crops rely upon honeybees for pollination, the combined value of which is now likely to exceed the $5 billion estimate calculated in 2014 (New Zealand Parliament, 2014).

Pastoral weeds, invertebrate plant pests, vertebrate pests, and micro-organisms are estimated to impose billions of dollars in costs on the New Zealand economy every year (Royal Society of New Zealand, 2014). Varroa is a particularly pernicious threat to honeybees in New Zealand¹¹ (Stahlmann-Brown et al. 2022). A report undertaken by the Office of the Auditor General at the time of the varroa incursion estimated that varroa would cost New Zealand between $400–$900 million over a 35 year period (Controller & Auditor-General of New Zealand, 2000), thus equating to an annualised cost of $11–$26 million (not inflation adjusted). This estimate seems to have been prescient when considered alongside our estimate of the value of over-winter colony loss effects, in which varroa causes a large proportion of winter losses.

Successful varroa control in autumn underpins colony survival over winter. If commercial beekeepers pursue rigorous varroa treatment in autumn, the costs will likely be more than recouped in the following honey/pollination season by reducing the greater consequential cost of colony loss. Moreover, successful varroa control has positive spillover effects for neighbouring colonies and apiaries.

We suggest the findings shown in this paper provide the basis of a business case for interested parties to invest in research to address all forms of colony loss and varroa in particular as the investment required can now be compared objectively to the potential long-term monetary gains.

Acknowledgements

We gratefully acknowledge the following people for their thoughtful input to this paper: Data and interpretation: Annette Carey, Ministry for Primary Industries. Pollination and area sown in arable crops: Keith Gundry, Pure Oil New Zealand Ltd; Ivan Lawrie, Foundation for Arable Research; Matt McCully, Midlands Apiaries. Honey volumes: Phil Edmonds, Apiculture New Zealand; Sean Goodwin, Pure New Zealand Honey. Honey prices: Six honey buyers who provided honey purchase prices on condition of anonymity. We also acknowledge Pam Booth, Manaaki Whenua – Landcare Research, for reviewing this paper and offering insightful feedback.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 As of June 2021, 5.7% of New Zealand beekeepers managed 85% of honey bee colonies according to annual disease returns filed with the New Zealand National American Foulbrood Pest Management Plan Management Agency. In contrast, 20% of beekeepers in Canada manage 80% of colonies (Canadian Honey Council 2022).

2 Biosecurity (National American Foulbrood Pest Management Plan) Order 1998.

3 For example, the average response rate in EU countries is around 5% (Gray et al., 2022). The share of colonies covered in the U.S. survey is estimated to be below 10% (Bee Informed Partnership, 2021).

4 To wit, MPI (2021) reported a price range for bulk mānuka honey of $8-$120 per kilogram for the 2020/21 season.

5 The price obtained for monofloral mānuka honey depends critically on the methylglyoxal (MGO) content or the Unique Mānuka Factor or UMF (a measure of unique signature compounds including leptosperin, DHA, and MGO).

6 The remaining colonies were used for activities that are not strongly affected by colony losses, e.g. queen rearing and the production of products such as beeswax.

7 $nF$ may also reflect the cost of cleaning brood frames that are not replaced.

8 See: https://www.mpi.govt.nz/food-business/honey-bee-products-processing-requirements/manuka-honey-testing/

9 As expected, these figures are all below the export prices for bulk export honey ($31 for monofloral mānuka, $17 for multifloral mānuka, and $12 for non-mānuka) reported by Statistics NZ Export Trade Data.

10 Beekeepers facing different costs can use Equations (1)–(3) to more accurately estimate the impact of over-winter losses on their own operations.

11 And indeed worldwide, having recently been found in Australia, the last large land mass without a significant varroa incursion. https://www.outbreak.gov.au/current-responses-to-outbreaks/varroa-mite