“It’s complicated”: Exploring preferences in small-scale forest management in Austria using multi-objective optimization

Technical Summary: "It's complicated": Exploring preferences in small-scale forest management in Austria using multi-objective optimization

Problem Statement
Small-scale private forest owners (PFO) manage a significant portion of Europe's forests and are critical for maintaining Forest Contributions to People (FCP). While literature characterizes these owners as multi-objective managers driven by non-commercial values (e.g., aesthetics, conservation) alongside material needs, a persistent discrepancy exists between their stated management objectives and their observed management behavior. Previous studies suggest that barriers such as lack of technical equipment, knowledge, time, and financial constraints prevent owners from acting on their stated goals. However, the specific mechanisms driving this "want-can" inconsistency remain under-explored, particularly regarding how perceived management complexity influences the translation of diverse values into actual forest composition.

Methodology
The study employs a novel methodological approach combining an exploratory survey with multi-objective, robust optimization (ROM) to analyze the forest management of small-scale PFO in Austria.

1. Survey Design: An online survey was conducted with 62 valid responses from Austrian small-scale PFO (properties <200 ha). The survey utilized the Nature's Contributions to People (NCP) framework to structure nine management objectives categorized into three groups:

   • Material FCP: Long-term income, meeting household needs, liquidity.
   • Regulating and Non-material FCP: Carbon storage/sequestration, natural hazards protection, ecological functions, general preferences.
   • Management Effort: Management complexity, management costs.
     Respondents rated the expected performance of six generic forest stand types (e.g., coniferous even-aged, mixed deciduous-coniferous uneven-aged, forests without intervention) against these objectives and allocated importance weights to the objectives.

2. Optimization Model: The study applied a multi-objective, robust optimization model (based on Knoke et al., 2015, 2016) to simulate an uncertainty-averse forest owner.

   • Input: The model used the average performance scores and standard errors derived from the survey respondents as input parameters.
   • Uncertainty Handling: The model incorporated an uncertainty set comprising optimistic and pessimistic performance estimates for each forest type across all objectives. A factor $m$ (ranging from 0 to 3) determined the level of uncertainty, with $m=3$ representing a high level of uncertainty (worst-case scenarios).
   • Objective Function: The model solved a MIN-MAX problem to minimize the worst underperformance ($\beta$) across all objectives and uncertainty scenarios, seeking a compromise solution that maximizes the guaranteed performance floor.

3. Analysis of Inconsistency:

   • Stated Objectives: Derived directly from the survey weights.
   • Observed Objectives: Inferred by running optimizations for various objective combinations and comparing the resulting optimal forest compositions against two baselines: the sample average (self-reported current composition) and the national average (derived from Austrian forest inventory data). The Bray-Curtis (BC) dissimilarity measure was used to identify which objective combinations produced forest compositions most similar to the observed baselines.

Key Results

• Stated Objectives: The survey revealed that small-scale PFO prioritize a mix of objectives, with "Long-term income," "Meeting household needs," and "Ecological functions" being the most relevant. The study found no strict dichotomy between material and non-material goals; owners perceive these as complementary rather than conflicting.
• Ideal Composition (The "Want"): When optimizing for the stated objectives (or all nine objectives simultaneously), the model consistently identified mixed deciduous-coniferous, uneven-aged forests as the dominant optimal stand type. This type was perceived as the most versatile, scoring highest across almost all material and regulating objectives. Pure deciduous stands were entirely excluded from optimal portfolios.
• Observed Composition (The "Can"): The actual forest compositions (both sample and national averages) were best explained by optimizations that prioritized reducing management complexity, reducing management costs, and enhancing liquidity.
• The "Want-Can" Inconsistency: A significant mismatch was identified. The forest composition that would best satisfy the owners' stated preferences (dominated by complex mixed forests) differs markedly from their current reality. The optimization results indicate that the current forest landscape is shaped not by the owners' primary stated goals (income, ecology), but by the constraint of management complexity.
• Role of Management Complexity: "Management complexity" emerged as the single most critical factor in explaining observed behavior. It is the only objective that consistently appeared in the top-performing objective groups that matched the baseline forest compositions. Paradoxically, owners ranked this objective as one of the least important in their stated preferences, viewing it as a constraint rather than a goal.
• Trade-offs: Achieving the ideal, multi-functional forest composition (high ecological and material performance) requires accepting higher perceived management complexity and lower liquidity/income in the short term. Conversely, the current observed behavior reflects a trade-off where complexity and costs are minimized, often at the expense of ecological and long-term income objectives.

Significance and Claims
The paper claims to offer a methodological advancement by using robust optimization to reverse-engineer "observed" management objectives from land-use patterns, thereby quantifying the gap between stated preferences and actual behavior.

• Rejection of Dichotomy: The study challenges the traditional "material needs vs. conservation" dichotomy, finding that small-scale PFO in Austria view these objectives as integrated and complementary.
• Complexity as the Primary Barrier: The central finding is that perceived management complexity is the key lever explaining the inconsistency between what owners want to achieve and what they actually do. The authors argue that owners do not lack the desire for multifunctional forests but are constrained by the perceived difficulty of managing them.
• Policy Implications: The paper suggests that forest conservation policies relying solely on profit-maximization incentives or assuming owners are purely profit-driven are insufficient. Instead, effective governance must address the "management complexity constraint." The authors propose that policy instruments should focus on reducing perceived complexity (e.g., through cost-sharing, in-kind payments, peer monitoring, and participatory design) to activate the potential of small-scale PFO to maintain and enhance FCP.
• Stewardship vs. Commodification: The findings support a shift in forest policy from a commodification logic to a stewardship approach that aligns with the relational values and plural motivations of small-scale owners.

Limitations
The authors acknowledge that the study is exploratory with a non-probability sample (78 initial responses, 62 valid), limiting generalizability. The static nature of the model cannot capture temporal feedback loops or gradual behavioral shifts. Additionally, grouping specific tree species into generic stand types may have introduced ambiguity in respondent perceptions. The results describe a phenomenon observed in the sample and should not be interpreted as prescriptive for all Austrian PFO without further representative validation.