Wage Determinant Factors for Farm-Support Paid Volunteers: Emerging Co-Creating Rural Tourism Addressing Labour Shortage in Rural Japan

Abstract

Volunteer tourism is garnering growing attention across various fields, allowing tourists to both consume and co-produce tourism services. In agriculture, however, this remains underexplored, despite a worsening farm labour shortage due to ageing populations and a lack of successors, particularly in industrialised nations. This issue threatens farm productivity and food security. This paper addresses this research gap by examining paid volunteer tourism platforms in Japan. It presents a framework highlighting the co-creation of local tourism demand and analyses wage determinants across 138 farms. Results show that corporate farms engaged in direct sales offer higher wages, especially when prices are elevated or locations are remote, suggesting wage premiums reflect labour shortages. Accommodation and Wi-Fi provision depend on farm finances and unused facilities. Organic and GAP-certified farms offer lower wages, likely due to higher production costs, despite producing value-added goods. As platform-based paid volunteer tourism meets the needs of both farmers and volunteers, its prevalence is expected to increase.

1. Introduction

A notable trend in rural tourism is its evolution towards directly benefitting both farmers and visitors through farm-based participation. This shift is largely driven by ageing and depopulation in rural communities, accompanied by a lack of successors, which has led to severe labour shortages in the agricultural sector, particularly in industrial countries. In this context, farm activities that engage volunteers to support elderly farmers or address workforce deficits are attracting growing attention, particularly among urban residents who combine rural tourism experiences with meaningful contributions.

With respect to volunteer tourism in agriculture, there are three types of volunteers, distinguished by the degree of compensation. The first type receives no payment at all, aligning with the conventional understanding of volunteer work. Traditionally, labour exchange has been practised among neighbouring farms to help manage peak labour demands; this reciprocal exchange functions as a form of non-monetary compensation. However, this practice has become increasingly difficult to sustain due to labour shortages and an ageing farm population.

The second type of volunteer does not receive wages but is provided with meals and accommodation by the host farm, as seen in programmes such as WWOOF (Working Weekends on Organic Farms). Additionally, student volunteering is also included in this category.

The third type receives wage payments, which we refer to here as “paid volunteers.” This category is expected to grow as farms seek to attract more helpers in response to deepening labour shortages, and as volunteers look for opportunities to travel to rural areas, gain farm experience, and partially offset travel costs through paid work. Despite its growing relevance, paid volunteer tourism has received little scholarly attention. This gap motivates our focus on this emerging form of volunteer tourism.

Terry (2014) highlighted the importance of volunteer tourism as a means of addressing labour shortages, focusing on its socio-economic implications [1]. Ando and Ohe (2016) examined the roles and benefits farm volunteers gain from participating in operations at suburban farms in Japan [2].

This emerging form of rural tourism differs from conventional models focused solely on consumption; instead, it emphasises participatory, paid volunteerism on farms and reflects a co-creative approach to agricultural production and other economic benefits to destination communities. It is especially appealing to urban dwellers seeking innovative ways to engage with rural communities and contribute to their sustainability.

To facilitate connections between farmers and interested volunteers, new matchmaking platforms have emerged that specialise in farm-based paid volunteer tourism. These platforms offer a novel structure within rural tourism, helping address the challenges of labour shortages while offering urban participants the opportunity to earn income through agricultural volunteer work.

Despite its growing relevance, there has been limited research on this emerging model, as is discussed later in the literature review. To bridge this research gap, the present study focuses on a paid volunteer platform operating within Japan’s farming sector.

First, it proposes a conceptual framework to capture the co-creative nature of paid volunteer tourism. Second, it investigates the determinants of farm wages—an area lacking national statistical data—in order to identify effective factors for developing this form of rural tourism. Finally, the paper offers policy recommendations aimed at supporting the growth and sustainability of farm volunteer tourism.

2. Literature Review

Volunteer tourism has gained increasing attention as an emerging area of study since the early 2000s [3,4,5,6]. McGehee (2014) conducted a comprehensive review of the field, identifying future research directions and emphasising the need for a unified and cohesive theoretical foundation [7]. Taplin et al. (2014) also reviewed earlier literature on volunteer tourism and presented an analytical framework aimed at enhancing the monitoring and evaluation of volunteer tourism programmes [8]. Grimm and Needham (2011) explored how destination and organisational attributes influence both volunteers’ and managers’ perceptions of volunteer motivations [9]. Hernandèz-Maskivker et al. (2018), from a managerial perspective, examined volunteer tourism’s impact on host communities and advocated transitioning from top-down volunteerism to a co-creative approach—an idea integrated into this paper’s conceptual framework, which is explained in the following section [10].

In the context of cross-border volunteer tourism, Keese (2011) assessed the criteria used by NGOs to select destinations and to discover how these organisations utilise tourist activities and place imagery to market locations and attract volunteers [11]. Grimm and Needham (2012) further examined promotional materials and found that prospective volunteers predominantly use the internet to identify opportunities [12].

Students are often considered ideal candidates for volunteer tourism, prompting both conceptual and empirical studies. Cunha et al. (2018) proposed a conceptual framework for student volunteering [13]. Konno et al. (2021) analysed managerial factors influencing student volunteer organisations [14], while Rodrigues and Abreu (2023) examined the role of universities in facilitating student volunteer tourism through co-creation experiential learning programmes [15].

In terms of disaster-related volunteerism, Fukui and Ohe (2017) analysed volunteer support in tsunami-affected regions of Japan using a time series model, highlighting the short-term role of volunteers prior to full-scale public restoration efforts [16]. However, their study did not address farm-based volunteerism.

Research on volunteerism within rural tourism contexts remains largely limited to the WWOOF domain, despite growing potential beyond WWOOF [17,18,19,20,21,22]. WWOOF, an international network, connects volunteer tourists with organic farms, providing opportunities for engagement in farming in exchange for meals and accommodation. It can be said that this farm volunteer programme set a good example for the development of farm volunteer programmes globally.

Urban agriculture has also been investigated, with emphasis on volunteer participation due to its proximity to consumers [2,23,24]. Dobernig and Stagl (2015) explored how urban food cultivation reflects a lifestyle blending consumption and production [25]. Tiraieyari and Krauss (2018), using hierarchical regression analysis, identified attitudes as the strongest predictor of Malaysian youth participation in voluntary urban agriculture programmes [26]. Kiptot and Franzel (2019) examined numerous donor-supported farmer-to-farmer extension (F2FE) initiatives in Kenya through a volunteer farmer–trainer model [27]. These examples demonstrate the multifaceted nature of volunteer tourism in agriculture, encompassing farmers, students, consumers, and international participants.

Tran and Tran (2024) explored the sustainability impacts of volunteer tourism in rural Vietnamese communities through semi-structured interviews, concluding that local community involvement in decision-making is critical [28]. In contrast, Zavitz and Butz (2011), studying an organic farming project in Costa Rica, found limited material contributions due to volunteer tourism being predominantly subordinate to tourism demands [29]. This finding suggests the balance between volunteer work and tourism behaviour. Ohe (2018, 2022) examined how variations in dairy farmers’ attitudes towards educational activities for schoolchildren and families significantly influenced managerial efficiency, not considering those of volunteers from outside the local community [30,31]. Let us summarise the results:

First, existing studies offer both affirmative and critical evaluations of volunteer tourism, cautioning against overly optimistic expectations and emphasising the need to examine the specific roles played by volunteers. Second, the concept of “co-creation” provides a valuable lens for advancing volunteer tourism and warrants integration into the development of its conceptual framework. Third, to the authors’ knowledge, no prior research has investigated nationwide platforms that connect volunteers with farms, particularly those involving paid volunteer models, from an economic standpoint. This study aims to address this gap.

3. Materials and Methods

3.1. Conceptual Framework

Paid volunteers differ from both conventional workers—who primarily seek income—and pure volunteers—who participate without monetary compensation. The motivations of paid volunteers encompass both farm assistance and leisure activities. The monetary component enhances their motivation, making them a distinct category of participants in rural labour markets. This study presents a conceptual framework in which farmers respond to labour shortages by engaging paid volunteers. We posit the existence of a paid volunteer market, which is a plausible assumption given the presence of privately operated platforms for paid volunteer tourism in this country. For analytical simplicity, we consider three primary actors: farmers, paid volunteer tourists, and the platform that facilitates their interaction.

The platform provides mutual benefits by connecting farmers with volunteers. For farmers, it offers an effective mechanism to alleviate labour shortages, thereby contributing to the sustainability of farm operations and local agricultural viability. For paid volunteers, it enables interactions with local communities, minor earnings to offset travel expenses, and opportunities to engage in tourism activities. By performing a matching function, the platform reduces search costs on both sides.

When farmers opt to utilise paid volunteers, we introduce a farm subjective equilibrium model to describe their decision-making process. In Figure 1, the horizontal axis represents volunteer labour input, while the vertical axis indicates value, ceteris paribus. Our analysis is grounded in standard microeconomic assumptions. The marginal benefit (MB) illustrates that farmers derive from paid volunteer labour.

For volunteers, participating in farm work helps subsidise their travel costs while enabling them to experience local tourism. Platforms specialising in paid volunteer tourism facilitate these matches. Although paid volunteers are legally entitled to at least the minimum wage, they may accept lower compensation than regular employees due to the additional utility derived from the tourism experience, based on the studies of behavioural and labour economics (Smith et al., 2022 [32]; Lavetti, 2023 [33]). This tourism externality would be composed of factors such as volunteers’ preference for the rural environment, environmentally friendly farming, and interest in rural revitalisation, etc. Farmers may offer higher wages in regions with more severe labour shortages.

The marginal cost (MC) curve reflects the expenses incurred in utilising paid volunteers—such as farm operation costs and accommodation or meal provisions. Typically, volunteers bear their own travel costs, while farmers may offer lodging and food depending on their individual policies. As volunteer activity increases, associated costs rise. Paid volunteers contribute to agricultural activities such as production, processing, and sales—particularly beneficial in rural areas facing acute labour constraints.

In the absence of a platform, the equilibrium wage rate (w₀) and labour input (On₀) are determined at the intersection of MB₀ and MC₀, marked as point e₀. At this equilibrium, total surplus is represented by area ade₀. Farmers pay w₀, with their surplus depicted by the upper rectangle (aw₀e₀), while volunteer surplus is shown by the lower rectangle (de₀w₀), reflecting their dual benefit of income and tourism engagement.

Introducing the platform yields two key effects. First, the MC curve shifts downward from MC₀ to MCpl due to reduced search costs. Second, the MB curve shifts upward from MB₀ to MBpl, driven by improved information and reduced uncertainty for the farm side to use paid volunteers. The new equilibrium, at point e₁, results in greater total surplus, represented by triangle bfe₁ (bfe₁ > ade₀).

Moreover, paid volunteer tourism has co-creative features in two domains. First, volunteers engage in collaborative agricultural work with farmers, reflecting supportive rather than purely economic motivations, unlike hired labour. Second, volunteers stimulate local tourism demand in the destination area. While their farm duties mirror those of hired labour, their underlying motivations generate co-creative dynamics that extend beyond the farm.

This co-creation effect is captured by the MBt curve, representing tourism externalities. The vertical distance between MBpl and MBt reflects the positive spillovers that volunteer activity generates for local tourism, which generates food consumption and economic benefits to the local economy. Farmers, by using the platform, indirectly contribute to this tourism demand, while volunteers directly foster it through their presence and participation.

Thus, the platform plays a central role in enabling this co-creation process between volunteers and farmers. While the surpluses themselves are not directly observable, this conceptual framework guides the empirical investigation into wage determinants as disclosed by the platform, from the farmers’ perspective.

3.2. Trend of Farm Volunteers

Although nationwide statistics on farm volunteers are not available, limited data have been released by the Tokyo Metropolitan Government, offering partial insights into farm volunteer trends within the prefecture. Despite the constraint, these data provide a useful approximation of broader patterns.

Figure 2 presents the trend in farm volunteer activity in Tokyo, depicting two key indicators: the number of newly registered volunteers and the number of dispatch cases. While registrations have steadily increased, the number of dispatch cases rose sharply after the COVID-19 pandemic, reaching nearly 400. This pattern suggests a growing demand for farm volunteers. The researchers identify both temporary and structural drivers behind this increase. Temporarily, demand surged due to the post-pandemic recovery of the food service industry, including restaurants, bars, and hotels. Structurally, the ageing farm population and ongoing depopulation of rural communities continue to intensify labour shortages.

Figure 3 details the age composition of newly registered volunteers from 2018 to 2020. A wide generational spread is evident, with individuals in their twenties registering most actively. However, registration does not necessarily translate into participation.

Figure 4 illustrates the age composition of actual programme participants, revealing a generational shift: older individuals in their fifties and sixties formed the majority in 2018, whereas younger cohorts dominated in 2020. Compared with the registration data, actual participants tend to be younger—primarily in their twenties to forties—which together account for more than half of the total. Due to the lack of more recent data, interpretations should be made cautiously.

3.3. Data and Methodology

The dataset was obtained from the website of Otetsutabi, a paid volunteer tourism platform established in 2018 by a young female entrepreneur. This platform has created a novel realm of volunteer tourism—one driven by a private initiative that had not existed before. The platform has already received several nationwide awards and attracted substantial public attention. It provides detailed information on the timing and nature of farm work, as well as the terms and conditions offered to paid volunteers. Thus, the platform connects farms with city residents who have a willingness to participate in volunteer tourism.

In principle, transportation and meal costs are not covered, although some farms do provide meals. Minimum wage standards are legally upheld. The survey period spanned from September to November 2022, encompassing data from 138 farms, of which 115 were corporate entities.

To investigate wage determinants, the researchers employed three models estimated via Ordinary Least Squares (OLS), an approach appropriate for the data structure—cross-sectional observations, continuous wage variables, and the absence of panel dimensions. To address the limited sample size and potential heteroscedasticity, they also applied bootstrap sampling, which yielded more statistically reliable results. Model 1 uses daily wages as the dependent variable, whereas Models 2 and 3 focus on total wages, each with a distinct set of explanatory variables. Model 1, based on daily wages, represents short-term labour pricing while Models 2 and 3, based on total wages, reflect the design of the overall volunteer package.

For Model 1, to evaluate the relationship between daily wages and production and direct selling activities, and volunteers’ competency, the explanatory variables include the following: main product categories: rice/wheat, peach, flower, and vegetable (binary: yes = 1, no = 0); engagement in direct sales (yes = 1, no = 0), serving as a proxy for earning capacity; and requirement for volunteers to hold a driving licence (yes = 1, no = 0).

Model 2 examines the relationship between total wages and fringe benefits, i.e., three service offerings provided to volunteers: accommodation service, meal service, and Wi-Fi service. The details are as follows: accommodation availability (five levels): individual room = 4, partially shared individual room = 3, individual or shared room = 2, shared room = 1, and day trip only (no accommodation) = 0; meal service availability (four levels): full board = 4, full board excluding weekends = 3, half board = 2, meal expenses subsidised or self-cooking supported = 1. The most common service was subsidy/self-cooking, with 62 cases, followed by half board, with 56 cases. Full board was uncommon. Wi-Fi availability (binary: yes = 1, no = 0).

Model 3 captures destination and farm-specific characteristics that may appeal to volunteers: population of destination municipality, distinctive farming practices, and an appealing point for volunteers. These are detailed as follows: population of the municipality (three levels): ≥500,000 residents = 2, ≥100,000 residents = 1, and <100,000 residents = 0; distinctive farming practices: organic and/or GAP (Good Agricultural Practice)-certified (yes = 1, no = 0); volunteer’s driving licence possession (yes = 1, no = 0).

Descriptive statistics for all variables are presented in

Table 1. Descriptive statistics of variables.

Variables	Sample Size	Mean	Std Err	Min	Max
Total wages (Yen)	138	44,195.2500	38,129.0100	2859	280,000
Daily wage (Yen)	138	5991.3700	1447.1380	2472	9000
Accommodation service (4: single, 3: single with shared area, 2: either simple or shared, 1: shared, 0: day tour)	138	1.9275	1.1692	0	4
Meal service (4: full board, 3: full board except weekend, 2: half board, 1: one meal or subsidy for cooking for yourself)	138	1.7754	0.888	1	4
Population of located municipality (2: ≥500 thousand, 1: ≥100 thousand, 0: <100 thousand)	138	0.3913	0.6881	0	2
Wi-Fi availability (yes = 1, no = 0)	138	0.6377	0.4824	0	1
Devising production process (yes = 1, no = 0)	138	0.2826	0.4519	0	1
Holding driving licence (yes = 1, no = 0)	138	0.0725	0.2602	0	1
Selling job (yes = 1, no = 0)	138	0.0870	0.2828	0	1
Product: rice or wheat (yes = 1, no = 0)	138	0.0435	0.2047	0	1
Product: peach (yes = 1, no = 0)	138	0.0435	0.2047	0	1
Product: flower (yes = 1, no = 0)	138	0.0362	0.1875	0	1
Product: vegetables (yes = 1, no = 0)	138	0.1014	0.3030	0	1

. The diagnostics indicate no irregularities, affirming the robustness of the dataset. Bootstrap techniques were employed, as noted, to strengthen inference given the modest sample size and variance concerns.

4. Results and Discussion

Table 2. Estimation result of model 1 (daily wage).

Explained Variable	Daily Wage
Explanatory Variable	Parameter	Z Value	VIF
Product: rice/wheat (yes = 1, no = 0)	1360.1120 ***	2.87	1.10
Product: peach (yes = 1, no = 0)	1248.5410 ***	2.89	1.13
Product: flower (yes = 1, no = 0)	1098.3260 ***	2.98	1.02
Product: vegetable (yes = 1, no = 0)	−849.5862 ***	−2.67	1.15
Direct selling (yes = 1, no = 0)	817.5552 **	2.09	1.17
Holding driving licence (yes = 1, no = 0)	831.2972 *	1.92	1.03
Constant	5793.0150 ***	39.32	-
Adjusted R2	0.1157
Sample size	138
Breusch–Pagan/Cook–Weisberg Test Chi2	2.68
Wald Chi2	30.15 ***

Source: Data was collected from Otetsutabi url. Notes: ***, **, * indicate 1%, 5%, and 10% significance, respectively. Variance Inflation Factors (VIFs) and heteroscedasticity tests were conducted using Ordinary Least Squares (OLS) estimation, followed by bootstrap sampling to enhance statistical robustness.

,

Table 3. Estimation results of model 2 (total wage).

Explained Variable	Total Wage
Explanatory Variable	Parameter	Z Value	VIF
Accommodation service (5 levels )	5052.9740	1.63	1.07
Meal service (4 levels )	−8282.2850 ***	−2.73	1.07
Wi-Fi availability (yes = 1, no = 0)	11,803.9000 *	1.83	1.01
Constant	41,632.3900 ***	6.45	-
Adjusted R2	0.0472
Sample size	138
Breusch–Pagan/Cook–Weisberg Test Chi2	20.94 ***
Wald Chi2	9.63 **

Source: Data was collected from Otetsutabi url. Notes: ***, **, * indicate 1%, 5%, and 10% significance, respectively. The treatments of VIFs, heteroscedasticity tests, and bootstrap sampling are aligned with those procedures in Table 2.

and

Table 4. Estimation result of model 3 (total wage).

Explained Variable	Total Wage
Explanatory Variable	Parameter	Z Value	VIF
Population (3 levels)	−11,439.5600 ***	−2.96	1.07
Holding driving licence (yes = 1, no = 0)	30,422.7900	1.28	1.03
Devising production process (yes = 1, no = 0)	−11,495.9800 **	−2.09	1.05
Constant	49,715.9100 ***	9.18	-
Adjusted R2	0.0900
Sample size	138
Breusch–Pagan/Cook–Weisberg Test Chi2	20.36 ***
Wald Chi2	13.19 ***

Source: Data was collected from Otetsutabi url. Notes: 1. ***, ** indicate 1%, 5% significance, respectively. The treatments of VIFs, heteroscedasticity tests, and bootstrap sampling are aligned with those procedures in Table 2.

present the results of the three estimation models. Variance Inflation Factors (VIFs) were consistently low—around unity—indicating no multicollinearity concerns. All three models are suitable for interpretation.

Table 2 shows the results of Model 1, which analyses daily wages. Each product-related parameter was statistically significant at the 1% level. The largest positive coefficient was observed for rice/wheat, followed by peach and flower. In contrast, the coefficient for vegetables was negative, suggesting that vegetable farms tend to offer lower wages due to their labour-intensive nature and higher demand for workers than that of other products, given the constraint of limited profitability. The production of rice and wheat is highly mechanised, and labour demand peaks during spring planting and autumn harvest. Compared with vegetables and flowers, the period during which labour is needed for rice production is relatively short. As a result, daily wages for rice and wheat work can be higher than those for other crops, given the short duration of peak labour demand.

Farms engaged in direct selling and volunteers with a driving licence exhibited positive coefficients, which is significant at the 5% and 10% levels, respectively. These results imply that direct selling increases farm revenue, enabling higher wage payments, while licenced volunteers contribute more effectively to operations. The results suggest that farms engaging in direct product sales are better positioned to offer higher daily wages, which may make paid volunteer work more appealing and activate further farm diversification.

Summary of Model 1: Direct selling plays an important role in increasing revenue and supporting higher daily wages. Volunteers who hold a valid driving licence tend to receive higher compensation, reflecting anticipated contributions and potentially offsetting travel and tourism-related expenses. Thus, farms that engage in direct selling may be particularly well-positioned to attract volunteers, thereby further strengthening their diversification efforts.

Table 3 reports the results of Model 2, focusing on total wage determinants related to farm-provided services. The accommodation service parameter was not statistically significant, reflecting its widespread availability (offered by 90.6% of farms). In contrast, meal service and Wi-Fi availability were significant at the 1% and 10% levels, respectively. Notably, the signs of the coefficients differed: meal service was negative, whereas Wi-Fi availability was positive. The negative meal service coefficient indicates that meal-related costs were likely deducted from the total wages offered.

Summary of Model 2: Accommodation did not influence wage levels due to its prevalence. However, meal service negatively affected total wages, mainly when volunteers prepared meals themselves or received subsidies. These findings suggest that accommodation, meals, and network connectivity are standard provisions factored into wage structures. Thus, fringe benefits do not have a positive impact on the total wages.

Table 4 provides the results of Model 3, which emphasises destination characteristics and production practices. The population variable had a negative and significant coefficient (1%), indicating that farms in less populated, remote areas tend to offer higher wages to attract volunteers. Interestingly, the variable of devising production processes such as organic farming and GAP certification was also negatively associated with total wages (5% significance), which is a notable result.

Although these practices typically command premium prices, they incur higher costs, and their benefits are not always reflected in domestic market prices—unlike European and North American contexts. Another factor is that volunteers are likely to favour these farming practices already, creating an externality that can justify offering lower wages. This externality reflects the volunteer’s willingness to forgo monetary compensation due to their intrinsic motivation to contribute to rural socio-economic and environmental development. The expected enjoyment of implementing tourism is also supposed to be included in the externality. Thus, volunteer tourism can be seen as providing benefits to both volunteers and farms. In particular, this form of rural tourism has the potential to become a promising model for rural development. Whereas traditional rural or agri-tourism typically involves urban tourists solely consuming tourism services in rural areas, paid volunteer tourism differs importantly. The platform creates opportunities for individuals who are motivated to engage in farm work, while simultaneously generating a market for intermediary services that connect volunteers with farms. In this context, we can say that this co-producing rural tourism could make a contribution to easing rural issues. The platform creates opportunities for individuals who are motivated to engage in farm work, while simultaneously generating a market for intermediary services that connect volunteers with farms. In this context, this co-producing form of rural tourism can be seen as contributing to the mitigation of rural challenges, and promotion is necessary to raise recognition of the emerging social roles of this type of rural tourism.

Summary of Model 3: Farms in smaller municipalities offer higher wages as an incentive. Organic and GAP-certified farms, despite their perceived market value, may face profitability constraints that suppress total wage offerings. The results are of interest as they show that Organic and GAP certifications are not positive on the total wages.

Collectively, the models reveal nuanced insights into wage determination.

Daily wages are influenced by product type, operational practices (e.g., direct selling), and volunteer attributes (e.g., driving licence). The results reflected the degree of product profitability and the extent of the labour shortage. Farms that engage in direct selling can offer higher wages and are often more attractive to volunteers. Thus, we can say that farm diversification will make farm activity not only more profitable but also attractive to volunteers. It can attract volunteers who are interested not only in agricultural production but also in business activities that involve interacting with consumers. In this context, farms pursuing diversification are better able to attract volunteers with broader perspectives.

Total wages reflect a broader range of factors—destination characteristics, farm practices, and service provisions. Negative wage determinants include meal costs and high-cost production systems like organic and GAP farming. The authors argue that these negative effects on total wages do not necessarily imply a disadvantage in attracting volunteers because paid volunteering has a co-creation nature containing externalities, as we can understand from the conceptual model. Thus, the farm can attract highly motivated individuals who are committed to sustainable farming practices.

Thus, platforms that facilitate paid volunteer tourism serve an effective function in recruiting individuals interested in rural heritage and agricultural engagement with wider perspectives. This function reduces the search cost for those with a volunteer mind. Volunteers value both fair compensation and meaningful tourism experiences. As shown in our analysis, farms engaged in direct selling tend to offer higher wages due to improved profitability. Enhancing farm business performance is, therefore, key to attracting volunteers through more competitive wages and services.

More broadly, given the ageing farm population and the rise in direct-to-consumer online sales, accelerated by COVID-19, demand for farm volunteers is likely to continue growing. Paid volunteers who combine agricultural work with local tourism represent a lifestyle model that bridges urban and rural experiences. The Ministry of Internal Affairs and Communication (MIC), Japan, encourages growth in the segment of “non-native individuals who maintain ongoing relationships with rural areas”—a profile that paid volunteers increasingly match. Thus, the platform can serve as a significant online mediator in this broader, long-term context, helping to cultivate rural–urban relationships in the post-COVID era. At the same time, it is important to keep in mind that volunteers’ interests are likely to become increasingly diverse in the future.

Regarding the empirical results, we need to account for potential sample selection bias in the estimation results, as the sample consists solely of farms registered on the platform. This limitation should be considered when interpreting the results.

Finally, platforms may foster a divergence between farms that successfully utilise volunteers and those that do not. This dynamic must be considered when designing recruitment strategies for paid volunteer programmes.

5. Conclusions

Farm volunteer tourism is gaining attention as a potential solution to Japan’s chronic agricultural labour shortages. This study addresses a relatively unexplored area—paid farm volunteer tourism—which presents a novel form of rural tourism bridging urban engagement with rural revitalisation. It aligns voluntary farm work with income-earning opportunities in rural areas. This study clarified the significance, potential, and considerations of volunteer tourism.

First, the researchers developed a conceptual framework incorporating subjective equilibrium, externalities, and co-creation perspectives. These dimensions have not previously been integrated into this topic. The framework illustrates how volunteers and farmers jointly generate positive externalities in local tourism demand through platform-mediated interactions. In other words, the platform acts as a catalyst for co-creation between both parties, enabling the emergence of destination-level tourism demand.

Second, empirical analysis using regression models—based on data from a Japanese platform—revealed several wage-determining factors. Farms that produce high-value crops and engage in direct sales tend to offer higher wages. Farms in less-populated municipalities also pay more, which is likely to offset severe labour shortages. In contrast, farms adopting organic or GAP-certified practices offer lower wages, likely due to elevated production costs not fully reflected in domestic market prices. Similarly, farms that provide meal services tend to deduct meal-related costs from total compensation, leading to lower wages. Provision of accommodation and Wi-Fi varies depending on farm resources and infrastructure.

Looking ahead, farm volunteer tourism is poised for growth, driven by increasing labour demand and the rise of direct online sales—trends accelerated by the COVID-19 pandemic. While the current volunteer pool is predominantly domestic, the resurgence of inbound tourism indicates future opportunities for international participation. Platforms will need to accommodate this shift by addressing cross-border volunteer demand and facilitating a more inclusive rural tourism market. However, this expansion may widen the disparity between farms that benefit from volunteer participation and those that do not, making user feedback from both volunteers and farms increasingly critical.

To prepare, farm operators must enhance their capacity to engage effectively with international volunteers. Capacity-building support and policy intervention should be considered to promote sustainable farm–volunteer partnerships. For individuals or farms not yet participating in the platform, support measures could help reduce the entry barriers.

The platform has the potential to foster new forms of rural–urban relationships in ways that have not previously existed and with a wider perspective. The case of paid volunteer tourism examined in this paper demonstrates this potential clearly, both within Japan and among international volunteer tourists, as mentioned above. In this context, the issue could be further explored through cross-country comparative studies. Moreover, the platform’s functions can reach far beyond this individual case and are likely to expand in scope as they increasingly contribute to alleviating rural issues as one of co-creation types of rural tourism. It is therefore important to continue monitoring how the platform evolves over time.

Finally, regarding the limitations of this study, several avenues remain for future research. As some influential data were not publicly available on the platform, complementary methods such as questionnaire surveys and in-depth case studies—targeting both farmers and volunteers—are needed. Further investigation into participant profiles, actual tourism behaviours, and the generation of external economies for local communities will also provide richer insights into the broader impacts of paid farm volunteer tourism. These socio-economic impacts of volunteer tourism warrant further investigation in future research.