Factors Affecting Activeness and Sustainability on Forestry in the Family Forests in Japan—From the Comparison between Aso in Japan and Styria in Austria

Abstract

Forestry in Japan and Austria share many similarities in their natural and social conditions. However, the Family Forest Owners (FFOs) in Japan seem not to be active and sustainable. To understand the factors affecting activeness and sustainability in family-owned forests in Japan, in 2021 and 2022, questionnaire surveys were done with members of Forest Owners’ Cooperatives (FOCs) in Aso, Japan, and Styria, Austria. Survey responses were comparatively analyzed via correlation analysis and binary logistic regression. Timber production was found to be more active in FOC Styria than in FOC Aso. One reason for this was the high-income dependence on forestry in Styria. Higher income was realized by self-harvest and the larger size of forest holdings and forest stands. The younger age of the members in FOC Styria, strongly affected by the pension system, leads to a higher self-harvest ratio. The culture of a sole child inheriting the family forest maintains the general size and scale of owned forests and stands in Styria. High distribution costs in FOC Aso reduced forestry income. As a result, sustainability was reduced in Aso because the availability of successors was low, and elderly forest owners who were once motivated by forestry tended to quit forestry.

1. Introduction

Forestry has become more significant because of global warming: wood products are renewable, associated with low greenhouse gas (GHG) emissions, and can store higher levels of carbon stock as well as in forests [1]. However, reforestation after harvest is required to maintain certain levels of carbon stock and timber supply. To implement this, active forest management is necessary.

In terms of timber supply, planted forests have been essential even in developed countries. The area of planted forests in Japan had started to increase since the 17th century and had almost doubled after World War II [2]. On the other hand, planted forests in Austria had been intensively developed in the 19th century. These two countries have a long history of forestry [3].

Concerning recent timber production, forests in Japan have an annual timber utilization ratio of only 23% (=harvested standing timber volume: 48 million m³/gross annual increment: 208 million m³ [4,5]), while in Austria the ratio is 89% [6]. Even though there are several similar conditions for forestry in both countries, the activeness of forestry is very different. As for similarities, both countries are characterized by the fact that small-scale forest owners occupy large parts of forest areas. In Japan, the low utilization ratio can be explained by the small size of forest holdings [7]. In this paper, we defined non-industrial, non-legal entities, and individual forest owners as “Family Forest Owners” (FFOs). The average holding size of FFOs in Japan is 6.7 ha [4], while in Austria the average size is 24.9 ha [8]. The holding size is larger in Austria than in Japan; however, small and fragmented ownership was also noted as a primary factor impeding the active utilization of forest resources in European countries [9,10,11].

Terrain conditions are also similar in these two countries. Steep terrain is an important factor in the inactivity of timber production in Japan [12]. In 42% of forests in Japan, the slope inclination is over 30 degrees [12]; in Austria, the figure is 28% [13]. Forests in Japan are steeper than those in Austria; however, both countries have steep terrain compared to other developed countries with active timber production [13].

Timber production is increasing in both countries: 18 million m³ in the 2000s to 27 million m³ in the 2010s in Japan, and 17 million m³ in the 2000s to 18 million m³ in the 2010s in Austria [14]. Most of the timber in Japan is supplied from planted forests (approximately 10 million ha, or about 41% of total forests [4]), while timber in Austria is supplied from commercial forest areas of 3.4 million ha (about 85% of total forests). The fact that timber production per ha was 2.7 m³ in Japan and 5.3 m³ in Austria demonstrates the high activity of forestry in Austria. However, log prices in 2021 were at a similar level: 131 euro/m³ (€1 = ¥130; the same applies below) for Japanese cedar and 110 euro/m³ for European spruce in Austria (since saw log price statistics in Japan show prices at the sawmill, which include transportation and distribution costs, a price of €10 was added to the log price on the road in Austria, an estimate modified from the average costs with information gathered interviewing logging companies and sawmills).

Despite the many similarities between the two countries, in Japan, there has been an increase in clear-cutting without planting, and about half of FFOs have expressed a desire to sell their forests [12,15]. Since replanting is essential to maintaining future timber supplies in Japan, this situation will affect the sustainability of its timber supply. FFOs in Japan are reducing their forests by 117,000 ha per year. Those forests have been sold to legal entities such as logging contractors or forest companies or have not been properly bequeathed to successors after the FFOs’ deaths. Conversely, private forest owners in Austria have increased their holding size [8].

In this study, we focused on FFOs, since FFOs produce most of the timber in Japan even though their share of planted forests is 29% (the shares of national forests, local governments, and private legal entities are 22%, 13%, and 36%, respectively) [4,16]. In Austria, meanwhile, small-scale forest owners occupy about half of all forest areas and contribute about half of timber production [17].

In 2018, we had a chance to obtain information about FFOs, timber supply chains, and support organizations of FFOs in Austria. The results of the interviews with FFOs showed that FFO forestry in Austria seemed to be quite active and sustainable compared to that in Japan. We then posed a research question: “What factors affect the low activity and sustainability of the management of family-owned forests in Japan”?

In this study, we defined “sustainability” as maintaining timber production in the future. To maintain timber production, forest holding size, skills and motivations for forestry, and appropriate successors should be maintained. Therefore, in addition to assessing the present situation of timber production, forestland transactions were compared between Japan and Austria.

2. Data and Methods

2.1. Framework

To answer the research question, we developed questionnaires collecting general information about FFOs and their forests. Questions addressed: (1) forest conditions; (2) FFOs’ management behaviors and attitudes; and (3) FFOs’ characteristics; these factors were considered to influence the activeness and sustainability of forestry.

The questionnaire was developed in 2020 based on one used in a previous study in Japan [18]. The previous questionnaire was originally developed for assessing thinning activity and conditions for deciding the final harvest. The questions were as follows: forest holding area; whether or not timber was produced within the previous 3 years and for what reasons; harvest volume; tending activities; future harvest plans; advisors for forest management; forest owners’ gender; age; place of residence; intention for future forestry; and income sources.

In this study, we focused on FFOs’ timber production, which represents the activeness of forestry. As many previous studies concerned with wood mobilization among FFOs referred to distance to the forests [19,20,21,22,23], we added questions regarding the location of owned forests and frequency of visits to owned forests to the questionnaire. In addition to the size of the forest holding, which will affect timber production [21,22,24,25], the number of dispersed forest holdings was added to analyze the fragmentation of forests, which affects forestry costs. We calculated the average size of forest stands by dividing the total forest area by the number of forest parcels.

To encourage the utilization of forest resources, Japan’s Forest Agency has implemented several policies that mainly target FFOs. Therefore, many FFO studies were done prior to 2000 [26,27]. In Japan, about 25% of timber was previously supplied by self-harvest [28], but FFOs have recently been selling their timber not by self-harvest but by stumpage sales, which have become especially popular among non-agricultural FFOs [29]. To analyze this effect, a question about the logging person was added to the questionnaire.

As mentioned above, FFOs in Japan were reducing their holding forest area; however, official data on forest land transactions were not obtained for the regions examined in the study. Therefore, forestland transactions were added to the questions in this study to assess the sustainability of FFOs. Questions were also added about whether or not a forest was purchased or sold within 5 years, its area, and the reasons for the transaction. Many studies have been conducted on forestland transactions in the US, where most forest owners have purchased some or all of their forest land [20,30,31]. However, studies on forestland transactions have been limited outside the US [32,33].

While the number of individuals who have acquired their forest ownership via inheritance or forestland transactions has been increasing worldwide, many recent studies have discussed nontraditional, nonagricultural, or passive forest owners [11,34,35,36,37]. We collected data from forest owners’ cooperatives (FOCs) because FOCs are the only institutions with data on regional forest owners in Japan. Thus, this study mainly focuses on traditional FFOs or their children’s generation. Although the Japanese government publishes agricultural census data, reports on forestry activities do not include forest owner attributes such as income sources, visit frequency, and forestland transactions [38]. Questions concerning the sustainability of FFOs’ forestry, the availability of successors, and the forestry skills of the children’s generation were added to the questionnaire.

To assess the factors that affected timber production or forestland purchase/sale, we used correlation analysis to examine the relationship between forest owners’ intentions and timber production or forestland purchase/sale. Responses to the question on forestland purchase or sale that explicitly provided numerical values for the purchased or sold forest area were scored as one. For the question regarding timber sales, responses stating that timber was sold in 2019 were scored as one. We asked about forestry results in 2019 in order to eliminate distortion since the timber market has been volatile between 2020 and 2023, largely due to the COVID-19 pandemic.

As with the explanatory variables, the multiple-choice answers, i.e., forest owners’ intentions (listed in Appendix A), were also scored as one. For FOC Styria, options related to recommendations were combined into one category: “recommendation from FOC, contractors, or relatives”.

The factors affecting timber production or forestland purchase/sale were assessed by binary logistic regression. Continuous explanatory variables were forest area, age, and number of residents, and dummy variables were forest location, timber production, and forest tending in 2019, more than half of timber was sold by the FOC member (mentioned as selling timber on consignment with the FOC), more than half of timber was harvested by family or employer (mentioned as self-harvest), visit frequency, boundary clarification, gender, main and secondary incomes, and the availability of successors. Data were analyzed using logistic regression with the forward stepwise method in IBM SPSS version 21. To confirm or support the results of the statistical analysis, we conducted the interview surveys in two regions in 2022 and 2023.

Many studies have emphasized the importance of FOCs [9,39,40,41]. FOCs in Japan and Austria are also important entities for FFOs because the share of timber supply through FOCs is approximately 20% in both countries. However, FOC membership has been decreasing recently in Japan but increasing in Austria. We focused on the differences in the timber supply systems in each FOC.

To assess the activeness and sustainability of FFOs’ forestry, it is quite significant to compare the conditions, institutions, and support organizations relating to forestry between the two countries. Concerning multinational studies, Triplat et al. [19] performed a comparative analysis of small-diameter tree supply using data from forest owners in four European countries and found differences in terms of what the owners valued. Focusing on “new owners” and “future owners”, Karppinen [34] reviewed findings from Finland and the United States, discussed differences between the two countries, and showed how qualitative studies on future owners can provide insights into the generation-bound values and objectives of forest ownership. Only a few studies have examined more than one country [42]. Leiter et al. [43] compared Japan and Austria from the viewpoint of forest resources; however, very few studies have compared the behaviors of FFOs between European and Asian countries.

2.2. Research Sites and Subjects

We selected Kumamoto Prefecture, Japan, for its active forestry and, within this prefecture, the FOC in Aso as our research object (hereafter “FOC Aso”), whose jurisdiction covers the northeastern part of the prefecture and has steep terrain (

Table 1. Overviews on FOC Aso and Styria.

	Aso	Styria
Ranking in log production	2nd/11 regions (Kumamoto pref.: 6th/47 pref.)	2nd/8 provinces (in 2020)
Production forest (ha)	68,000	860,000
Members	5751	15,670
Members’ forests (ha)	51,000	323,000
Average holding size (ha)	8.9	20.6

Data sources: Interview from FOC Aso and Styria, BMLRT (2021) [17].

).

Regardless of holding size, forest owners have voluntarily joined FOC Aso, but most of them are small to medium FFOs. The FOC federation in Kumamoto Prefecture consists of 15 regional FOCs, including Aso (a national federation that has also been established by a group of prefectural federations). After World War II, FOCs in Japan supported the planting and tending of FFOs as points of contact for subsidy payments or as labor providers. They have also assisted in the harvesting and log marketing of FFOs.

In Aso, timber harvesting is conducted using chainsaws, excavators with winches, processors, and forwarders. Traditional cable yarding is occasionally performed at steep and large harvesting sites. Generally, felled trees are pulled by cable winches on skid roads; logs are then produced using processors and transported using forwarders to timber yards beside forest roads. The steep terrain allows only forestry machines to move on skid roads.

In Austria, meanwhile, we selected the FOC federation in Styria (Waldverband Steiermark; hereafter “FOC Styria”) as the other research subject because its jurisdiction also covers a region with active forestry and has steep terrain. In this region, timber harvesting is generally performed using chainsaws, agricultural tractors with winches (trailers), or tower yards, typically in steep terrain. As in Aso, these cable yarding systems are costlier than a vehicle logging system that uses harvesters and forwarders on gentle slopes.

FOC Styria manages mainly small- to medium-scale private forests. This FOC consists of 10 regional FOCs, each of which is composed of several subordinate FOCs. Small to medium FFOs freely join these subordinate FOCs, which automatically makes them members of FOC Styria. FOC Styria was established in 1990 with the strong support of the Styrian Chamber of Agriculture in the implementation of the joint marketing of logs supplied by FFOs. Thus, the placement of FOC Aso in Kumamoto Prefecture is similar to that of the regional FOC in Styria.

2.3. Questionnaire and Field Surveys

We conducted a preliminary survey in Aso in November 2020. The interviews with the staff of FOC Aso and a non-profit organization for assisting FFOs were done to obtain general information about FFOs and the timber supply chain and to give an overview of FOC Aso. We then discussed with the FOC Aso staff the questions and the presentation method of the questionnaire survey to its membership. Because FOC Aso members do not use the Internet for communications, we mailed our survey to them. Budget limitations caused us to send questionnaires to only 20% of the members of FOC Aso, who were randomly sampled; a total of 1155 questionnaires were mailed in December 2020.

This study adopted a mainly quantitative approach to examining forest owners’ behaviors; however, interview surveys were necessary to confirm the questionnaire results. We also conducted semi-structured interviews with 5 members and 3 staff members of FOC Aso in November 2023. We asked interviewees about the status of forest property and timber sales, the timing and trigger of inheritance or forestland transactions, and legal or cultural limitations for forestland inheritance and transactions.

In Styria, we conducted a preliminary survey in November 2019 to obtain general information about FFOs, the supply chain of timber, and an overview of FOC Styria. The interviews were conducted with the officer of the Styrian State Forest Department, staff members of FOC Styria and the Association of logging contractors, 2 logging companies, 3 FFOs, and 2 sawmills. We also consulted with the FOC Styria staff about the questions (which were translated into German) and the survey method. We had been exploring opportunities for a pretest interview survey but could not implement it because of the travel ban imposed by the COVID-19 pandemic. Due to the tight schedule, we skipped the preliminary study, and the questionnaire survey was delayed until 2022. As discussed below, a preliminary survey should have been conducted remotely.

Because subordinate FOCs had a membership of around 300, it was difficult to choose appropriate subjects. In addition, as this was our first attempt, we could not predict the response rate. As recommended by the FOC Styria staff, we conducted the survey online with 8700 FOC Styria members with registered Internet addresses to maximize the number of responses and minimize survey costs. Appendix A lists the questions in both surveys. To confirm the questionnaire results, in November 2022, we conducted semi-structured interviews with 11 members of FOC Styria, 4 regional FOC staff in Styria, and a staff member of the chamber of agriculture. We asked FOC Styria to select members who recently inherited, bought, or sold their forests. However, we could not meet with FFOs who recently sold forests. We asked the interviewees about the status of forest property and timber sales, the timing and trigger of inheritance or forestland transactions, and legal or cultural limitations for forestland inheritance and transactions.

For FOC Aso, 10% of the mailed questionnaires did not reach its members because its address database was outdated. We received 335 responses (29% response rate), of which only 288 were valid (

Table 2. Overview of the results of questionnaire surveys.

		Aso	Styria
		n = 288	n = 1479
Average size of forest property (ha)		16	28
Average size of forest parcel (ha)		3.6	17
Purchase forest > 0 ha (%)		6.9	15
Sell forest > 0 ha (%)		4.5	2.4
Sell timber in 2019 (%)		18	67
Harvest by family or employer (%)		4.5	90
Tending area > 0 ha (%)		24	70
Average age		70	49
Gender (%)	Male	88	90
	Female	16	10
Average number of residents		2.6	4.0
Availability of successor (%)		42	84
Main income (%)	Agriculture	24	29
	Forestry	2.8	7.2
	Self-employed	6.9	6.4
	Salary	20	37
	Pension	32	9.5
	Rent	2.1	0.7
	Dividend	0.0	0.2
	Other	0.3	0.9

). For FOC Styria, we obtained 1800 responses (21% response rate), of which 1479 were valid. Because most respondents in both FOCs are heads of household (81% in Aso and 91% in Styria), we used their age to represent the age of forest owners.

2.4. Overview of the Results of Questionnaire Surveys

2.4.1. Property Size and Location

As shown in Table 2, the average forest areas in Aso and Styria were 16 and 28 ha, respectively. The average holding size in FOC Aso is more than two times larger than the national FFO average. However, it was smaller than that in FOC Styria. Furthermore, the average size of forest stands in Aso and Styria is 3.6 and 17 ha, respectively. These size differences were significant.

Excluding windbreaks, forest properties are rarely adjacent to the residences or farmlands of forest owners in Aso. Therefore, we did not include the option “close to residence” in the FOC Aso questionnaire. Seventy-four percent of FOC Aso answered that their forests were “located in the same municipality”. Eighty-two percent of FOC Styria answered that the location of their forests was “close to residence”.

Forest location is closely related to the frequency of owners visiting the forest. In Styria and Aso, 77% and 8% of respondents visited their forests once or more per month, respectively. This frequency reflects the forest owners’ activities in forestry, as shown in Section 3.1.2.

In Japan, many forest owners do not know the exact boundaries of their land assets because they inherited forests without helping to manage them, or many forest owners are not identified in the land register because of address changes or failure to comply with inheritance procedures. Nationally, 46% of forests in Japan have been delineated by cadastral surveys [44]; however, according to FOC staff and forest owners in Aso, the boundaries have in most cases been set mainly based on the memories of forest owners or people from the same community. Approximately 80% of respondents in both FOCs knew the boundaries of their forests, which is consistent with the result of the interviews: most forest owners in Aso used to be motivated to conduct forestry work.

2.4.2. The Age of Forest Owners

The average ages of FOC members in Aso and Styria were 70 and 49 years, respectively. Seventy years old was quite older than the average age in Aso regions, 53 years old [45]. The Aso region is rather aged compared to Japan’s national average age of 46 years old. On the other hand, forty-nine years old was similar to the average age in Styria, 44 years old, while Austria’s national average was 47 years old [46]. The population in Aso is about 10 years older than in Styria. However, the difference in the average age of respondents was greater than 20 years, and the difference was statistically significant. Additionally, the percentages of respondents over 60 years old in Aso and Styria were 91% and 18%, respectively.

Unfortunately, as the age distribution of FOC members was not available, we could not check the bias caused by questionnaire methods. The average age of respondents in FOC Aso was similar to that of previous studies [28,29,47]. However, there was some possibility that more elderly members tended to respond in Aso. Conversely, it is also possible that in Styria, younger members tend to respond due to the online nature of the survey.

The average age of interviewed FFOs in Styria was under 40 years; thus, they were much younger than the average questionnaire respondents. This suggests that the heirs tend to be younger when they inherit forests. Five interviewed FFOs inherited their forest properties from their parent or partner who became eligible to collect their old-age pensions. The Agricultural and Forestry Pension Act (AFP: Bauern-Sozialversicherungsgesetz) stipulates that pension recipients must transfer their agricultural and forestland as a condition for receiving the pension. Four interviewed FFOs inherited their forestland because the age of the previous generation was considered to be indirectly connected with AFP.

Residents of Japan between 20 and 59 years of age, including foreign residents, are required to enroll in the National Pension (NP) system. Moreover, residents who work in an office or factory should enroll in the Employees’ Pension Insurance (EPI) system. The public pension system consists of two tiers: the NP covers all people, and the EPI additionally covers company or public-sector employees. Residents who make a living from agriculture can enroll in the Agricultural Pension (AP) system, which additionally covers agricultural workers. The AP of Japan is similar to the AFP of Austria, although it only focuses on agricultural land [48].

On the other hand, according to the staff of FOC Aso, inheritance of forestland generally occurs after FFO owners’ death. As a result, FFO owners in Aso would inherit their forestland when they were around 55 years old. As the average life-span of a male in Japan was 81 years old, the average number of years remaining for a 55-year-old man in Japan was 27 [49].

Many respondents in Aso were pensioners. In contrast, pensioners comprised only 9% of the respondents in Styria, where most forest owners are salaried. The primary income of 27% and 36% of respondents in Aso and Styria, respectively, comes from agriculture or forestry. To determine the type of forest owners [11,36], taking secondary income into account, traditional forest-owning farmers who earned their primary income from agriculture and a secondary income from forestry, or vice versa, were considered to be 2.1% in Aso and 24% in Styria. The percentages of the members who earned their secondary income from forestry were 11% and 51% in Aso and Styria, respectively. These results show that the income sources of the two FOCs are different. In the interview survey, many of the respondents in Aso were found to be formerly traditional farmers who had stopped farming due to age.

Pension payments in Japan begin once the recipient turns 65 years old, and forest owners’ pensions are not restricted by holding forest properties. NP payments, which total about 510 euros per month, are not enough to live on. As a result, most pension recipients without additional EPI or AP payments need to have other sources of income.

3. Results

3.1. Timber Production

3.1.1. Analysis of Descriptive Statistics on Timber Sales

A large difference in the activeness of timber production between the two FOCs was found: the proportions of members in Aso and Styria who sold timber in 2019 were 18% and 67%, respectively.

A factor that might affect this difference was the ratio of self-harvest; the ratio of respondents in FOC Aso was 28%, which is over three times lower than that in FOC Styria (90%). The age of FFOs is thought to affect the differences in the self-harvest ratio. Because age had a significantly negative correlation with self-logging in both FOCs (the coefficient was –0.095 in Aso and –0.102 in Styria, p < 0.05). The results of interviews also showed the difference between the regions: all FFOs can self-harvest their forests in Styria, while one FFO can self-harvest in Aso.

The lower self-harvest ratio in Aso resulted in lower income from timber, which is thought to affect the lower activities of FFOs. Most of the respondents in Aso receive a low stumpage price because they outsource harvesting work and timber marketing to FOC or logging companies. Their stumpage price was paid after subtracting the logging, and log distribution costs from the log price. Stumpage price in Japan is a small part of the log price; for example, the national average stumpage price of Japanese cedar was 25 euro/m³ compared to the sawlog price of 131 euro/m³ in 2021 [50].

Table 3. Correlations between the reasons for selling timber and timber productions.

	Necessity of Management	Merchan- Table Age/Size	Need Income	Recommendation
				FOC	Contractor
Aso	N.A.	0.545 **	0.255 **	0.265 **	0.145 *
Styria	0.741 **	0.430 **	0.315 **	0.175 **	0.026
	Price increase	Available road	Self-consume	Damage	Others
Aso	0.069	0.305 **	0.160 **	0.184 **	0.250 **
Styria	0.162 **	0.150 **	0.484 **	0.471 **	0.064 *

Note: * and ** denote significance at the 5% and 1% levels, respectively.

shows the results of the correlation analysis on selling timber. Tree size and the need for income affected the respondents’ decision to sell timber in both regions. Recommendations from the FOC also influenced respondents’ decisions, although recommendations from logging contractors also had a significant effect on Aso. This reflected the fact that logging contractors in Aso sometimes paid a higher stumpage price than FOCs because of their lower marketing costs.

Most respondents in FOC Styria use a joint marketing service. In contrast, only 14% of the respondents in FOC Aso used this type of marketing service. Since it was not consistent with the results of the interview survey, this result may be due to respondents’ misunderstanding of our question. Our question was “how do you sell your timber?”, which respondents in Aso may have interpreted to mean “where did you sell your timber in 2019?”. Based on this assumption, we estimate that the percentage of FOC Aso that uses a marketing service is close to 77% (= 40/52). In both regions, FOCs were major contributors to their members’ timber production. At any rate, that misunderstanding could be avoided by pre-test.

In both regions, forest owners’ decision to sell timber was influenced by self-sufficiency and forest damage. The coefficients for these variables were higher in Styria. This is consistent with the fact that biomass energy utilization is quite active and that climate-related or bark beetle damage is frequent in Styria. The presence of roads in respondents’ forests affected timber sales strongly in Aso because of its lower density of forest roads.

Forest owners in Styria were asked about the necessity for forest management. This was the most affected intention for the decision on timber sales. For forest owners in Styria, managing and tending forests for sustainable use is linked to forestry income.

3.1.2. Analysis of Binary Logistic Regressions on Timber Sales

On a binary logistic regression, the probability P of timber sales was predicted in relation to the independent variables x_i. The regression equation is as follows:

$$ln\left(P|1 - P\right)=B_0+B_1{x}_1+B_2{x}_2+\cdots +B_n{x}_n$$

(1)

where B₀ is a constant and B_i are coefficients for the significant variables; “Sell timber by FOC”, “Self-harvest”, “Tending”, “Boundary: Map and site”, “Boundary: FOC”, “Sell forest” in the case of Aso (

Table 4. Results of binary logistic analysis on timber sales.

FOC	Variable	B	p	Exp(B)
Aso	Sell timber by FOC	4.90	0.000	134
	Self-harvest	3.58	0.002	35.8
	Tending	1.14	0.042	3.12
	Boundary: Map and site	2.80	0.005	16.5
	Boundary: FOC	3.90	0.008	49.4
	Sell forest	2.07	0.017	7.93
	Constant	−5.64	0.000	0.004
	Percentage correct			93.3
Styria	Forest size	0.036	0.000	1.04
	Number of parcels	0.067	0.014	1.07
	Near residence	0.362	0.026	1.44
	Self-harvest	0.461	0.023	1.59
	Tending	0.594	0.000	1.81
	Once a week	0.605	0.001	1.83
	Once a month	0.411	0.006	1.51
	Number of residents	0.085	0.035	1.09
	Main income: Pension	−0.478	0.022	0.620
	Secondary income: forestry	0.671	0.000	1.96
	Constant	−2.14	0.000	0.118
	Percentage correct			72.7

).

The exponential of B_i, Exp(B_i), represents the ratio change in the odds of the event of timber sales for a one-unit change in predictor x_i. For example, Exp(B) for the variable “Sell timber by FOC” is equal to 134 in Aso. This means that the probability that a respondent sold timber by FOC was 134 times larger than the probability that the respondent did not sell timber by FOC, all other things being equal.

This suggests that FOC Aso plays a critical role in the decision making of the members’ harvesting and log sales. “Boundary: FOC” also showed quite large odds. This result is related to the fact that many forest owners entrust the management of their forests to the FOC through the Forest Planning System.

In this model, the p-values for each variable, which was the probability of the null hypothesis “coefficient of the variable is equal to zero”, were almost all smaller than 0.05, showing that the listed variables were statistically significant. The p-values for the omnibus test were smaller than 0.05, showing that this model was statistically significant. This model for Aso also had a high correctly classified rate of 93.3%, meaning that this regression model predicted probability P properly. The later models in this study also had higher correctly classified ratios than 72.7%.

“Self-harvest” showed quite high odds for Aso. Forest owners in Aso who knew their forest boundary properly tended to sell timber. Additionally, “Tending” showed relatively high odds. These results suggest that respondents in Aso who are motivated to manage their forests actively harvest forests and sell timber.

Respondents who sold forests had a high probability of selling timber in Aso. According to the staff of FOC Aso, FFOs who were used to being motivated by forestry tend to quit forest management. In Styria, proximity to the forests, self-harvest, tending activity, and income from forestry showed relatively large odds.

3.2. Forestland Purchases

3.2.1. Analysis of Descriptive Statistics on Forestland Purchases

In Aso, most of the increases in forest properties were due to inheritance. The number of respondents who purchased or inherited their forestland was 20 and 26 individuals, respectively; however, the total area of inherited forest is 429 ha, which is four times larger than that of purchased forest (93 ha). In contrast, most of the increases in forest properties in Styria were the result of purchases. The number of respondents who purchased or inherited their forestland was 220 and 75, respectively, and the total area of purchased forest was 1.449 ha, twice that of inherited forest (747 ha).

The proportions of respondents in Aso and Styria who purchased their forestlands were 7.0% and 15%, respectively, and the ratios of the total purchased area to the total respondents’ forest area in Aso and Styria were 2.2% and 3.5%, respectively. We do not consider these percentages to be small because they do not include non-respondent members and nonmembers of FOCs.

The results of the correlation analysis between the respondents’ intentions and forestland purchases are shown in

Table 5. Correlations between the intentions for purchasing forestlands and forestland purchases.

	Expand Forestry	Enlarge Property	Development	Nontimber Use	Recommendation
					FOC	Contractor	Relatives	Others
Aso	0.470 **	0.421 **	0.000	0.241 **	0.151 *	−0.023	0.446 **	0.084
Styria	0.421 **	0.270 **	0.090 **	0.091 **	0.134 **			0.231 **

Note: * and ** denote significance at the 5% and 1% levels, respectively.

. The purchasing behaviors of respondents in both FOCs are influenced by their desire to expand management or property and by recommendations from FOCs or relatives. Relatives in Aso had a larger impact on forest purchases. “Others” showed a positive effect in Styria because many respondents stated that they purchased forests for energy self-sufficiency or to acquire the neighboring forestland.

3.2.2. Analysis of Binary Logistic Regressions on Forestland Purchases

Table 6. Results of binary logistic analysis on forest purchases.

FOC	Variable	B	p	Exp(B)
Aso	Sell timber	1.71	0.001	5.55
	Once in several years	−2.07	0.055	0.127
	Number of residents	0.361	0.007	1.43
	Main income: Pension	1.11	0.049	3.04
	Constant	−4.31	0.000	0.013
	Percentage correct			92.5
Styria	Number of parcels	0.149	0.000	1.160
	Near residence	0.571	0.009	1.771
	Within the district	0.711	0.000	2.036
	Within the province	0.784	0.002	2.190
	Main income: Dividend	2.418	0.054	11.226
	Presence of inheritors	0.487	0.036	1.628
	Constant	−3.455	0.000	0.032
	Percentage correct			85.3

presents the results of the binary logistic regression analysis on forest purchases. In Aso, “Sell timber” shows quite a high probability of buying forestland. Respondents who mainly depended on their pensions had a higher probability of buying forestland. Respondents who did not visit their forests frequently were less likely to purchase forests.

In Styria, “Sell timber” was not significant. Respondents whose forests were located near their residences showed a high probability of purchasing forests. The respondents who had inheritors showed a high probability of purchasing forests. These results are consistent with the interview results. However, respondents whose forests were located within a district or province had also a high probability of purchasing forests.

Respondents who depend mainly on dividends have a high probability of purchasing forests. This reflected that wealthy owners expanded their forest lands. Concerning this, the staff of FOC Styria mentioned two trends. One was that wealthy owners expanded their forest lands to obtain hunting licenses. Because a forest owner with at least 115 hectares of contiguous huntable forestland is able to hunt wild animals in that forest [51]. Another was that wealthy owners are actively investing in timberland in Styria.

3.3. Sales of Forestland

3.3.1. Analysis of Descriptive Statistics on Forestland Sales

Decreases in the area of forest properties in Aso were due to both living donations to their descendants and sales to others. The number of respondents in Aso who sold and donated areas of forestland were 13 and 10, respectively, and the total forest areas sold and donated were 22 and 35 ha, respectively. In contrast, the area of forest properties in Styria decreased mostly through living donations. The numbers of respondents who sold and donated forestland were 34 and 22, respectively, and the total forest areas sold and donated were 169 and 1.107 ha, respectively.

The proportion of members in Aso and Styria who sold forest areas was 4.5% and 2.3%, respectively; however, the ratios on an area basis were 0.5% and 0.4%, respectively, which were lower than the ratios of the total purchased area to the total respondents’ forest area.

From the interview survey, the difference between regions was confirmed: purchase was passive in Aso but active in Styria. FFOs in FOC Aso said that they would purchase forestlands only if the conditions were quite good, while they often asked to purchase forestlands. On the other hand, most of the interviewees indicated that they would like to expand their forestlands if the adjacent forests were to be sold at a reasonable price.

Table 7. Correlations between the intentions for selling forestlands and forestland sales.

	Unprofitable	Need Income	No Inheritor	Recommendation
				FOC	Contractor	Relatives	Others
Aso	0.003	0.106	0.131 *	0.355 **	0.271 **	0.099	0.250 **
Styria	0.115 **	0.087 **	N.A.	0.050	−0.006	0.309 **	0.592 **

Note: * and ** denote significance at the 5% and 1% levels, respectively.

shows the results of the correlation analysis on forestland sales. In Aso, selling behaviors are highly influenced by recommendations from FOCs or logging contractors. Therefore, the sales activities of those forestry enterprises that aim to secure the standing timber harvesting business are related to the sales of forestlands. This trend in FOC Aso is consistent with the results of a Japanese government survey that showed that 21% of forest management entities, including FFOs, wanted to quit their business, and 17% wanted to downsize their business [15].

3.3.2. Analysis of Binary Logistic Regressions on Forestland Sales

In Styria, forest sales are also highly influenced by recommendations from relatives. The coefficient of the variable “Others” which includes “attractive price”, “payment for inheritors”, and “development under public works”, positively affected forestland sales.

Table 8. Results of binary logistic analysis on forest sales.

FOC	Variable	B	p	Exp(B)
Aso	Sell timber	1.66	0.007	5.27
	Age	0.082	0.024	1.09
	Constant	−9.54	0.001	0.000
	Percentage correct			95.3
Styria	Number of parcels	0.074	0.020	1.08
	Within the district	1.45	0.000	4.25
	Once in several years	3.17	0.012	23.8
	Number of residents	0.139	0.044	1.15
	Secondary income: forestry	−1.24	0.007	0.291
	Constant	−4.98	0.000	0.007
	Percentage correct			97.8

shows the results of the binary logistic regression. In Aso, respondents who sold timber had a high probability of selling forests. The age of the forest owner significantly affected selling behavior: older forest owners in Aso were more likely to sell forestland. The results are consistent with the results of interviews with the staff of FOC Aso, who found that once-motivated FFOs for forestry tend to sell their forests and quit forest management.

In Styria, forest owners who rarely visited their forests or who had their forests “Within the district” tended to sell forests. These results are consistent with the results of interviews in Styria, where FFOs whose forests were located far from their residences tended to think that their forests were difficult to manage. Forest owners whose secondary income comes from forestry tend to be motivated to manage their forests, and the probability that they will sell their forests is low.

4. Discussion

4.1. Framework of the Analysis

In this study, we selected research sites where timber production is active. Because differences in the backgrounds could be found when we selected active forestry regions. This hypothesis was almost correct. However, different results could be found when we selected inactive forestry regions.

We tried to ask the same questions of members of both FOCs; however, options were slightly different in a few questions because of the different backgrounds or distribution systems. Even though the options were the same, interpretations of words could be different. For example, the “availability of successors” was often not decided in Styria, mainly because the children of FFOs are too young to decide. However, it was not decided in Aso because the children of FFOs are not interested in forestry. According to the results of the two regions’ interviews, if the successors were decided, they would live near their forest properties after they inherit forests in Styria, but that would not necessarily be the case in Aso.

Concerning the methods of questionnaire surveys, by mail in Aso and by Internet in Styria, age bias may have occurred. The response rate in Aso was slightly lower than in previous studies using the same survey method in Japan [25,47]. This might reflect FFOs’ lower motivation for forestry in Aso, but it was at a similar level compared with previous studies in Western countries [19,24]. Due to the lack of pretests, misunderstandings of questions have occurred in Aso. Some of these problems were corrected by the interview surveys. In any case, motivated forest owners who live near their forest holdings would tend to respond to the questionnaire in both regions.

In Styria, we could obtain a certain number of responses; however, the response rate was lower than that of previous studies. The Internet survey was selected for the respondents’ convenience; however, it seemed not to contribute to any increase in responses. Additionally, it was possible that younger people familiar with the Internet tend to answer Internet-based surveys.

To understand the cultural and legislative backgrounds, interview surveys are essential. It was also useful to confirm the results of the questionnaires. However, there are limitations to obtaining sufficient and unbiased data.

Concerning questions or options that were asked of FOC members, academic background and occupation were not selected in order to minimize the number of questions as much as possible. If we could have obtained that data, further analysis could have been done about FFOs whose first income was not derived from agriculture or forestry but whose secondary income was derived from forestry.

4.2. Factors Affecting Timber Production

Timber production was found to be more active in FOC Styria than in FOC Aso. The fact that a certain amount of income from forestry is an important income source in Styria affected the difference.

The size of forest holdings is critical for annual timber production [47,52] and for earning sufficient income for living from forestry. This was the case in Styria but was not observed in Aso. Respondents in Aso could not earn sufficient income from forestry because of the small size of forest holdings and lower stumpage prices.

General awareness of inheritance was found to affect the size of forest holdings in Styria because all of the interviewees inherited their previous generation’s lands whole. They mentioned that their siblings had waived their rights to the land or that they paid their siblings according to the assessed value of the land. On the other hand, the staff of FOC Aso mentioned that FFOs with larger forest holdings were more prone to disputes over inheritance.

Weiss et al. [10] reported on the restrictions on buying or selling agricultural and forest land in Austria. For example, under the Land Transaction Act (LT: Grundverkehrsgesetz), someone other than the siblings or children of forest owners can purchase agricultural and forest lands only when the commission for land transactions grants its approval and neighboring farmers waive their priority to buy land for sale. Additionally, the Austrian Forest Act (AF: Forstgesetz) prohibits the subdivision of forest parcels into parcels that are too small for regular management. This legislation might indirectly maintain the size of forest holdings and forest stands by avoiding fragmentation of forest holdings or the transfer of forest holdings to unmotivated persons.

By leading to higher forestry income, self-harvest positively affected timber production, especially in Aso. The aging of forest owners in Japan is related to the decrease in the ratio of self-harvest, which is consistent with the ratio of stumpage sales in Kyusyu increasing with the age of forest owners [29]. Additionally, respondents in Aso who self-harvest their trees tend to be younger than those who do not. Therefore, aging appears to be indirectly related to the decrease in the probability of timber sales in Aso. Shigematsu et al. [23] and Poje et al. [22] also showed that elderly forest owners are less likely to harvest.

The average age of forest owners in Finland, Slovenia, Spain, Sweden, and the USA is about 61 years old [16,30]. Therefore, respondents in Styria (average age: 49 years old) are younger than the above results. It is likely that the pension system in Austria explains why forest owners in Austria are younger than those in these other countries.

According to previous studies in Japan [29,47], the average age of forest owners was above 70 years old. In Aso, the average age of respondents was also 70 years old, which is similar to previous studies in Japan. The age of FFOs in Japan would be older than that in Western countries.

Lindroos et al. [53] also reported that self-employed NIPF owners tend to be significantly younger than those who are not self-employed, and NIPF owners who self-harvest tend to live close to their forests and farms, while NIPF owners who only do tending do not live close to their forests and do not farm.

Proximity to the forests tends to increase the probability of timber sales in Styria, which is consistent with previous studies [53,54]. However, this was not the case in Aso, where the probability of timber sales was influenced by boundary clarification. This legal risk for harvesting could be reduced when FFOs live near their forests and visit them often, which reflects the indirect relationship between proximity to the forests and boundary clarity.

According to Conway et al. [24], absentee landowners are unlikely to harvest timber. In this study, absenteeism was not shown to significantly affect timber production because there were few absentee forest owners among the respondents in both FOCs.

Synergism between agriculture and forestry [51,55] is demonstrated by forest-owning farmers who divert their tractors to be used for agriculture in the summer and for forestry in the winter. This applies to Styria, where livestock production is popular, and 29% of respondents earned their primary income from agriculture. However, 39% of respondents in Styria whose secondary income came from forestry earned their main income from a salary. This shows that income sources other than agriculture and forestry are also important to FFOs.

On the other hand, that synergism does not apply to Aso, where paddy field cultivation is popular and farmers have small tractors (around 30 PS) only for agriculture. Pensions were the primary source of income for most Aso respondents. Additionally, selling forestland strongly increased the probability of timber sales in Aso. This suggests that aged forest owners who lost the motivation for forestry tend to quit forestry and earn an additional income from the sale of both trees and forest land. While EPI generally pays 1730 euros per month to elderly couples, NP pays only 1020 euros at best [56]. Therefore, beneficiaries of NP will need additional income.

Previous studies have reported the positive impact of FOCs on timber production [10,35,39,41]. The FOCs in both regions facilitated timber sales through the joint marketing services of timber sales. However, the types of services offered in each region differed. On the one hand, the headquarters of FOC Styria establishes its selling prices and volumes through consultation with the timber industry and the regional FOC staff. Then, the staff in regional FOCs coordinates the harvest volume with members, and the logistics for transporting harvested logs along forest-side roads are passed directly to timber companies. Additionally, FOC members in Styria were relatively independent of the FOC because 12% of members of FOC Styria sold their timber directly to sawmills.

On the other hand, FOC Aso delivered harvested logs to the two log auction markets that the FOC also operates. The staff of timber companies and log distributors gather in these markets twice a month to purchase logs. This log distribution system is very popular, especially in the western part of Japan. Such a market is significant for high-quality logs needed to produce lumber for interiors or furniture because the bidding price would be higher than general logs for construction. However, distribution costs are increased by the sorting, piling, and secondary transport of logs from markets to timber industries. Since most of the logs produced by FOC Aso were general logs, these additional costs lower forestry income in this region.

4.3. Factors Affecting Forestland Purchases

A high percentage of forest owners, especially in Styria (15%), purchased their forests. The motivation for forestry strongly influenced the decision to purchase forests, as the main reason for purchasing forests in both regions was to expand forestry. Respondents in Aso who sold timber were very likely to have purchased forests. Although forestry income had no significant effect, respondents whose main income was a pension were highly likely to purchase forest. This indicates that elderly people tend to acquire forest land and sell standing timber to supplement their pensions.

Meanwhile, in Styria, forest owners whose main source of income was dividends were very likely to purchase forests, and their forests were located outside of their residential district. According to Kuvarda [35], nonagricultural forest owners do not depend on timber revenues; rather, “their economic interests mainly focus on achieving a positive cash flow”. According to the results of interviews with staff of FOC Styria, there are two possibilities; wealthy owners expanding their forest lands to obtain hunting licenses or investing in timberland. Concerning this, Hogl et al. [36] mentioned that forest owners who are “Small-towners” with a rural background tend to acquire their forests by either purchase or barter, and they also tend to lease their forest areas.

In Aso, forest owners who seldom visit their forests have a low probability of purchasing forests. In Styria, respondents who own forests close to their residences tend to purchase forests. Both show that proximity to the forests affects the probability of purchasing forestlands. Motivation for purchasing neighboring forests was not high in Aso but quite positive in Styria.

4.4. Factors Which Affect Forestland Sales

Forestland sales were not more active than purchases in both regions. The respondents who sold timber showed a high probability of selling forests in Aso. Conway et al. [24] showed that the debt-to-income ratio is positively related to the decision to sell forests. Moreover, Young et al. [52] showed that the income from forest harvests has a significant impact on commercial harvests. Consistent with these results, some FOC members in Aso sold both standing timber and forest land to supplement their pensions. The sale of standing timber or forest land to logging contractors raises concerns about the sustainability of timber production because logging contractors generally do not plant trees after clear-cutting.

The high-income dependency on forestry showed the low probability of selling forests in Styria. Motivated forest owners in Styria seldom sell their forestland.

Older forest owners in Aso were more likely to sell their forests, which is consistent with the results of Markowski-Lindsaya et al. [30], who found that the increasing age of FFOs increases the probability of selling or giving away their wooded land. This was indirectly related to the lack of heirs in Japan.

FOC members in Styria who lived far from their forests sold their forests. This is consistent with the results of Markowski-Lindsaya et al. [30], who found that “Having a home within 1 mile of wooded land increases the probability of being unlikely to sell or give away wooded land”.

5. Conclusions

Timber production was found to be more active in FOC Styria than in FOC Aso. The higher income dependence on forestry affects the difference in activeness. FOC members in Styria depended on forestry income, but FOC members in Aso depended on pensions. The need for money was a positive driver of timber production in both regions; however, the income seems to be used for daily living in Styria but as a supplement in Aso. The higher self-harvest ratio and larger size of forest holdings and forest stands in Styria lead to sufficient income for living from forestry.

The differences between the size of forest holdings and forest stands were affected by the general awareness among the members of FOC Styria that forests should be inherited by a single child. The differences were also affected by the strong willingness of the members of FOC Styria to purchase neighboring forests.

The arrival of an era of mass inheritance may accelerate the fragmentation of forest parcels in Japan, where about 30% of the population is over 65 years old. To reduce or maintain forestry costs, the general awareness that forests should be inherited by a single child or merged several forest stands into large parcels should be fostered. The introduction of policies on land transactions in Japan, such as the LT Act and the AF Act, would support this.

Respondents in the two FOCs differed significantly in their average age. The pension system in Austria, such as the AFP, favors an earlier time for forest owners to bequeath and inherit their forests. Thus, the younger age of respondents in Styria was associated with its higher self-harvest ratio as opposed to FOC Aso. In Aso, the practice of self-harvest can be increased by transferring forest ownership to young and motivated FFOs, an act that would require strong policy measures similar to those of the Austrian AFP.

FOCs play important roles in both regions in increasing timber production and log marketing. However, log distribution costs were higher in FOC Aso than in Styria. If FOC Aso directly delivers its members’ logs from harvesting sites to timber factories, it could reduce distribution costs.

Selling forestland strongly increased timber production in Aso. Additionally, forest owner age positively affected timber production in Aso. These suggest that elderly forest owners who lost the motivation for forestry tend to earn income from selling timber and forests and quit forestry. From the viewpoint of sustainability, the situation in Aso seems to be the reproduction of timber production on a reduced scale.

This study also observed that members who were motivated by forestry purchased forestlands in both FOCs. Transferring forestland from inactive forest owners to motivated ones leads to increased timber production in Styria. On the other hand, the sustainability of timber production in Aso, where some motivated FFOs passively purchased forestland, seems to be lower because the ratio of availability of successors to harvesting skills of successors was low.

Since conditions surrounding FFOs vary from region to region, different results will be obtained from FOCs in other regions. Therefore, it was difficult to arrange similar conditions among different regions and countries, for example, forest holding size. In addition to that, different cultural and economic backgrounds are also limitations of comparative studies.

To understand the behaviors of diverse forest owners, general trends must be assessed among forest owners who are not members of FOCs or other forest-owning enterprises. However, obtaining sufficient and unbiased data in this regard would require substantial effort and budget. Besides quantitative analysis, future studies will benefit from performing qualitative analysis as well.