Serbia’s industrial sector ended 2025 with growth, but only narrowly so. The headline figure from the February 2026 edition of MAT – Macroeconomic Analyses and Trends is that total industrial production increased by just 0.9% over the course of the year, a result that kept industry in positive territory but fell materially below the momentum many expected at the start of the year.
That number matters because it captures far more than a simple slowdown. It reflects the convergence of several pressures acting on Serbia’s productive base at the same time: weaker external industrial demand from the eurozone, especially Germany and Italy; climate-driven volatility in electricity generation; geopolitical risk transmitted through the energy sector; and a domestic manufacturing structure in which a relatively small number of branches carry a disproportionate share of output and export growth. What emerges from the MAT analysis is not a story of generalized collapse, but one of narrowing industrial breadth. Serbia did not lose industrial capacity in a broad-based sense in 2025, but it did become more dependent on a handful of sectors and more vulnerable to shocks in strategically important branches.
The deceleration became especially visible at the end of the year. In December 2025, total industrial production was 5.7% lower than in the same month of the previous year. This year-end decline was severe enough to reshape perceptions of the full-year trend and to highlight how fragile the annual growth outcome really was. The December contraction was not evenly spread across all segments of industry. Mining still recorded year-on-year growth of 4.4%, and the sector of electricity, gas, steam, and air conditioning also remained slightly positive with growth of 0.7%. The real weakness came from manufacturing, where output fell by 8.3% year-on-year in December, making it the decisive drag on the aggregate result.
At the annual level, the industrial picture looks more balanced but still weak. Manufacturing output rose 1.1% in 2025, mining grew by 4.7%, while the electricity, gas, steam, and air-conditioning sector declined by 1.8%. Since manufacturing is the dominant industrial segment, its low growth rate meant that even relatively solid mining performance could not lift the entire industrial sector much above stagnation. This becomes even clearer when looking at breadth: only 12 out of 29 industrial branches recorded a rise in physical output during the year. In other words, less than half of Serbia’s industrial branches contributed positively to growth.
That distribution is one of the most important signals in the report. The issue is not only that industrial growth slowed to 0.9%. It is that the number of branches carrying industry forward was limited. When a country’s industrial result depends on a narrow set of outperformers, the overall system becomes more sensitive to interruptions, whether those interruptions come from weather, sanctions, changing demand in export markets, or firm-specific problems. Serbia’s 2025 industrial result fits that pattern closely.
The MAT report places strong emphasis on the change in dynamics during the closing months of the year. According to the analysis, the full-year slowdown reflected both the external environment and a sharp shift in domestic conditions late in the year. Global economic uncertainty, geopolitical risk, and weak European industrial conditions reduced the room for a stronger rebound. But the authors also argue that the final result was decisively shaped by the deterioration in the oil refining segment, where slower operations and eventual stoppages at the Pančevo refinery effectively consumed much of the expected industrial growth. Without that disruption, the report suggests, the annual result would likely have come close to plan and may even have exceeded it.
This is a crucial point for understanding the 0.9% number. It was not simply the outcome of a broad cyclical cooling across all branches. It was also the product of concentrated weakness in one strategically important area. The branch “production of coke and petroleum products” experienced an extraordinary collapse at year-end. In December 2025, output in this branch fell by 94.3% year-on-year, and the production index relative to the 2024 average fell to just 6.5. The scale of this decline was so large that the MAT authors state plainly that this one branch, considered in isolation, can explain the entire fall of manufacturing output in December.
The significance of that statement goes beyond the refinery itself. It reveals how concentrated Serbia’s industrial vulnerability has become. When one branch can shift the direction of total manufacturing, it means the system is operating with insufficient balance. In a more diversified industrial structure, weakness in one large branch would still matter, but it would not dominate the annual outcome to this degree. Serbia’s 2025 result therefore reveals a structural issue: industrial performance is being driven by too few positive engines and can be derailed by too few negative shocks.
At the same time, there were genuine industrial success stories. Mining was the only major sector that managed to remain above the average 2024 level throughout all 12 months of 2025, and all mining branches recorded year-on-year growth in December. This included metal ore extraction, coal extraction, crude oil extraction, and other mining activities. Even here, however, the report notes that the longer upward cycle that had begun in the middle of 2023 appears to have weakened from June 2025 onward. So while mining remained positive, the trend was not one of accelerating industrial support but of decelerating resilience.
The energy supply segment also deserves closer attention. The sector of electricity, gas, steam, and air conditioning accounts for 15.3% of Serbia’s industrial production, making it too important to be treated as a peripheral factor. According to MAT, after a pronounced downward trend that lasted from December 2023, the sector finally began to stabilize from the middle of 2025 onward. This stabilization was achieved thanks to higher production of thermal electricity from public power plants and growth in solar-based generation. The main obstacle remained hydropower. Drought conditions pushed hydropower production down sharply from April 2025 onward, and it remained significantly below both the previous year and the medium-term average until the end of the year. Only in November and December did a modest recovery begin. Over the whole of 2025, hydropower production was down approximately 18.5% year-on-year. In January 2026, however, output recovered by 11.1%, supported by heavier rain and snow in Serbia, the wider region, and Central Europe.
That pattern is important because it shows that even where industrial stabilization was occurring, it was fragile and partially weather-dependent. Serbia’s energy-industrial system in 2025 was therefore hit by two distinct forces: a hydrological shock affecting electricity production and a geopolitical shock affecting oil refining. These were different in origin, but both affected industrial statistics and both fed into the subdued annual growth rate.
The manufacturing side of the story is even more revealing. While annual growth of 1.1% in manufacturing seems modestly positive, the internal composition of that result shows strong divergence between sectors. The most visible outperformer was the branch “production of motor vehicles and trailers,” which benefited from the start of production of the electric Fiat Grande Panda in Kragujevac. According to the report, average monthly gains in output during the first ten months of the year were around 4%, and by the end of 2025 the level of production in the sector was approximately 60% higher than the 2024 average. The contribution of this branch to manufacturing growth was enormous. In December, it contributed 4 percentage points to year-on-year manufacturing growth, and over the full year it contributed 1.8 percentage points to the manufacturing sector’s total growth of 1.1%. That means that without automotive production, manufacturing would not have grown at all.
The rubber and plastics sector was another strong performer. In December 2025, output in this branch rose by 7.5% year-on-year, contributing 0.62 percentage points to manufacturing growth. Across the full year, production in rubber and plastics increased by 16.6%, contributing 1.33 percentage points to manufacturing growth. This sector therefore joined automotive production as one of the two critical industrial pillars supporting manufacturing performance in a difficult year.
What these data show is that Serbia’s industrial system in 2025 did not so much broaden as rebalance around a narrower set of winners. Automotive and rubber-plastics production effectively compensated for major weakness in oil refining and softness across several other branches. That is a workable short-term model, but it is not a particularly robust one. If a future shock were to hit automotive demand or a major component exporter, Serbia’s aggregate industrial growth could weaken even more sharply.
The food industry illustrates the opposite dynamic. It remains one of the largest pillars of manufacturing, accounting for 19.8% of total manufacturing output, yet its performance in 2025 was weak. Output in food production declined by approximately 1.5% over the year. The report notes that after a brief improvement in November, when production rose 2.1% year-on-year, output fell again in December, down 3.2%. The longer-term trend in food production has been on a downward path since the middle of 2024, and by the end of 2025 the level of production was comparable to that seen in August 2023. Within food production, the sharpest December contraction came from fruit and vegetable processing, where activity fell by nearly one-third, or 31.9%.
That weakness matters because food processing is large enough to influence total manufacturing dynamics materially. If such a major branch is flat or falling while growth is concentrated in automotive and rubber-plastics manufacturing, then the industrial structure becomes more exposed to external demand conditions and less anchored in broader domestic processing capacity.
The report also offers an important perspective through the lens of industrial-use groups and technological complexity. In 2025, the only stable long-term growth pattern was observed in capital goods production. At the same time, across end-use categories, energy production declined by 9.5%, durable consumer goods by 5.6%, and non-durable consumer goods by 2.9%, while intermediate goods excluding energy rose by 5.7% and capital goods rose by 7.7%. This composition suggests an industrial model increasingly tied to the production of components, machinery, and investment-linked goods rather than broad-based consumer manufacturing.
The technology breakdown reinforces this interpretation. The full positive contribution to manufacturing growth in 2025 came from sectors classified as medium technological complexity, with medium-high and medium-low technology groups together contributing 1.8 percentage points. High-technology production declined by 2.5%, while low-technology production fell by 2.1%. Serbia’s industrial growth in 2025 was therefore not driven by a transition into a more advanced high-tech profile. It was driven by stronger performance in medium-technology segments embedded in European industrial supply chains.
This raises a strategic question for 2026 and beyond. A growth model centered on medium-technology manufacturing can deliver exports and employment, but it does not automatically generate higher domestic value capture, deeper innovation capacity, or resilience against external cyclical downturns. If Serbia remains concentrated in medium-technology supply-chain roles while its largest export markets in Europe remain weak, then even good firm-level performance in a few sectors may not translate into strong aggregate industrial growth.
The external environment described in the report supports that concern. The eurozone industrial cycle remained soft through the beginning of 2026. The manufacturing PMI was 49.5 for the EU, 49.1 for Germany, and 48.1 for Italy in January 2026, all below the 50 threshold separating expansion from contraction. MAT characterizes the German industrial problem as structural rather than cyclical. That is a serious warning for Serbia, given how deeply the country is tied to German and broader EU supply chains. If Germany is not simply in a temporary slowdown but in a longer period of industrial adjustment, then Serbia’s industrial strategy may need rethinking.
This context makes the 0.9% growth figure even more revealing. It is not merely a disappointing annual result. It is a signal that Serbia’s industrial model has reached a more constrained phase. Growth is still possible, but it now depends on careful navigation of multiple risks: external demand weakness, energy security, weather volatility, narrow export concentration, and the technological ceiling of a medium-complexity manufacturing structure.
The immediate implication for 2026 is that normalization in oil refining and continued momentum in automotive production could lift industrial growth somewhat above the weak 2025 level. But the broader implication is more strategic. Serbia needs not just industrial expansion, but wider industrial breadth. It needs more than one or two strong branches to carry manufacturing. It needs a more diversified structure in which food processing, electrical equipment, chemicals, machinery, and energy-linked industries can all contribute more consistently. Otherwise, future annual results may continue to depend on whether a small number of sectors outperform strongly enough to offset concentrated shocks elsewhere.
The 2025 outcome therefore stands as both a warning and a diagnostic. Serbia preserved positive industrial growth, but only at 0.9%, and only because a few sectors were strong enough to compensate for severe disruptions in others. The result is not catastrophic, but neither is it reassuring. It marks the point at which industrial growth has clearly become more conditional, more uneven, and more dependent on strategic choices that go well beyond simple volume expansion.
