Spain’s Road Freight Sector Mapped by Interviews, Statistics and Seasonal Forecasts

Road freight remains the backbone of inland goods movement in Spain, but a new open-access study in European Transport Research Review argues that the sector can only be understood by combining three lenses at once: what industry insiders perceive as its strengths and weaknesses, how sharply its commodity segments actually differ in the official statistics, and what a seasonal forecasting model can and cannot promise about the near future. Drawing on 103 expert interviews, quarterly and monthly data from Spain’s National Statistics Institute, and a SARIMA model, the research paints a picture of a structurally strong yet persistently strained logistics system whose 2025 outlook is far more conditional than headline growth figures might suggest.

The qualitative strand rests on 103 semi-structured interviews conducted between January and April 2025 with senior managers, company owners, fleet and operations managers, and representatives of drivers’ and sectoral associations. Participants were recruited through professional networks using purposive sampling with quotas by role and firm size, a design the authors stress captures a broad range of informed perspectives rather than a statistically representative sample. Interview responses were thematically coded and consolidated into a SWOT matrix, with inter-coder consistency checks through iterative review of the codebook.

Across the interviews, consensus emerged on the density of Spain’s road infrastructure, the operational flexibility of road haulage, and the growing role of GPS, telematics and digital visibility tools in routing and cargo traceability. On the weakness side, the most frequently repeated constraints were diesel dependence, congestion in major metropolitan and logistics corridors, deteriorating or insufficient secondary infrastructure, and above all the shortage of professional drivers. Respondents described the labour issue not as a simple recruitment problem but as a structural bottleneck affecting scheduling, route coverage and the ability to absorb demand peaks. Opportunities centred on digitalization, fleet-management automation, lower-emission vehicles and e-commerce expansion, while the dominant threats were rising fuel and energy prices, macroeconomic volatility, tighter environmental regulation, competition from foreign operators with lower labour costs, and the gradual policy push toward modal shift. Notably, there was more agreement on bottlenecks than on solutions.

The quantitative strand begins with commodity-group comparisons based on the quarterly Permanent Survey on Road Freight Transport, covering the first quarter of 2002 to the third quarter of 2024. The researchers aggregated the Eurostat/UNECE NST 2007 nomenclature into six categories: agricultural products and live animals; food products and fodder; petroleum products; minerals and construction materials; machinery, vehicles, manufactured goods and special transactions; and other goods. Each group contains 91 quarterly observations, and because these are time ordered, the authors interpret their Welch two-sample t-tests as exploratory long-run comparisons rather than causal estimates, with Bonferroni correction setting the adjusted significance threshold at 0.0033 across 15 pairwise tests.

The results confirm that the Spanish freight market is anything but homogeneous. All 15 pairwise mean comparisons remained statistically significant after Bonferroni adjustment, and Hedges’ g effect sizes ranged in absolute value from 0.75 to 5.96, with fourteen comparisons showing large effects and only the food-and-forage versus machinery-and-vehicles comparison showing a medium effect. Minerals and construction materials stood out as by far the largest segment by mean transported volume and also the most volatile, with the highest standard deviation and a coefficient of variation of 0.501, compared with 0.248 for food products and forage and just 0.127 for other goods. Petroleum products remained the smallest segment by average volume.

The temporal profiles reinforce this segmentation. Minerals and construction materials showed the sharpest expansion and contraction over the two-decade series, confirming the strong historical linkage between freight intensity and construction activity in Spain, while agricultural and food-related goods displayed smoother trajectories consistent with relatively inelastic demand for essential goods. Machinery, vehicles and manufactured goods followed a more gradual upward recovery reflecting industrial reactivation, and positive skewness in the construction-related segment indicates episodic surges above its central tendency, making exceptional peaks most likely in precisely the segment that is already the largest.

For forecasting, the study turned to the monthly aggregate series from January 2012 to November 2024, modelling 155 observations with the Box-Jenkins methodology. After transformation and differencing, the selected specification was a parsimonious SARIMA (0,1,1)(1,1,0)12 model, chosen because it minimized both the Akaike and Bayesian information criteria among candidate specifications while maintaining significant parameters and an interpretable structure. An Augmented Dickey-Fuller test rejected the null of a unit root after differencing, and the ARCH test remained marginally above the conventional 5% threshold, providing no strong evidence of conditional heteroscedasticity.

The diagnostics, however, impose serious limits on interpretation. On the transformed and differenced estimation scale, the model’s reported R-squared of 0.248794 means the fitted component captures approximately 24.9% of observed variability, leaving roughly 75.1% unexplained. Ljung-Box tests revealed statistically significant residual autocorrelation at several lags, particularly 3 to 10, 12 and 24 to 29, indicating that short-run and longer seasonal dependence persists in the residuals. The authors are explicit that this residual structure reduces forecast reliability and may cause uncertainty to be understated if the model were treated as fully white noise, and that the specification is retained only as a conditional seasonal benchmark, not a fully validated structural forecasting model.

The conditional 2025 baseline forecasts for June through November reproduce the recurrent seasonal pattern and yield point values approximately 4.0% above the corresponding 2024 total for those months, with the largest absolute difference in October. Yet the study warns against reading this arithmetic increase as evidence of sustained growth. Because 2024 already exceeded pre-COVID levels in the aggregate series, the univariate model cannot distinguish post-pandemic normalization from a new structural trend, and the qualitative and commodity evidence suggests the aggregate baseline may conceal offsetting sector-specific trajectories, from comparatively stable agri-food flows to pronounced cyclical uncertainty in construction materials and potential structural headwinds for petroleum-related freight during the energy transition.

For planners, the authors recommend treating the forecasts as a seasonal scenario rather than a deterministic expectation, useful for identifying months when fleet availability and driver scheduling may become more demanding and lower-pressure windows suitable for preventive maintenance, but always updated with current information on industrial production, construction activity, fuel prices and regulation. The study’s limitations are candidly stated: a purposive rather than representative interview sample, exploratory time-ordered commodity comparisons, and a model excluding exogenous variables. Future work, the authors suggest, should estimate sector-specific, multivariate and hybrid models incorporating industrial production, construction activity, energy prices and regulatory drivers, so that the structural constraints identified by operators can be connected more directly to the demand dynamics visible in the data.

Subject of Research: A mixed-methods study of Spanish road freight transport combining SWOT interviews, commodity-group statistics and SARIMA forecasting from 2012 to 2024.

Article Title: Road freight transport in Spain, 2012–2024: a mixed-methods assessment with SARIMA forecasts

Article References: Dancausa Millán, M. G., & Millán Vázquez, M. G. (2026). Road freight transport in Spain, 2012–2024: a mixed-methods assessment with SARIMA forecasts. European Transport Research Review, 18(1), Article 88. https://doi.org/10.1186/s12544-026-00850-0

Image Credits: AI Generated

DOI: 10.1186/s12544-026-00850-0

Keywords: road freight, Spain, SARIMA forecasting, SWOT analysis, commodity groups, driver shortages, logistics, transport policy, time series, INE data, seasonality, supply chain