A recession is a broad decline in economic activity that can show up in production, consumption and income. A commonly used shorthand is technical recession: 2 consecutive quarters in which gross domestic product (GDP) is lower than in the previous quarter. That threshold is useful, but it is not a complete diagnosis of an economy.
Romania met this condition at the start of 2026. In the data available at the end of September, GDP stopped falling from one quarter to the next but had yet to grow. A company planning its last quarter needs to distinguish that national picture from demand in its own market.
TL;DR. According to Eurostat’s revised data, Romania’s GDP fell by 1.9% in Q4 2025 and by 0.1% in Q1 2026 compared with the previous quarter, so the country went through a technical recession. In Q2 2026 the change was 0.0%, and compared with last year the economy is 2.0% smaller, the weakest annual change since Q4 2020.
Industry, consumption, distribution and purchasing power point in the same direction, with a few exceptions worth watching. This article brings the figures together, with a chart and source for each, then sets out the questions a company should ask before adapting its offer and budget.
The charts use official data series available when this article was prepared. Each has a “How it was calculated, and the source” box with the institution, table and calculation. Statistical releases can later be revised, so the figures are a dated snapshot rather than a live reading.
What does recession mean in Romania?
In Romania, as elsewhere, real GDP measures economic output after accounting for price changes. The technical recession rule checks whether it fell for 2 consecutive quarters. The quarterly comparison uses seasonally adjusted data, so recurring seasonal changes do not distort it.
The technical definition can be checked against figures, but it has a limit. The technical threshold does not say how deep the fall is. An economy can enter technical recession with a drop of a few tenths of a percent. Another can fall by several percent and leave the definition after a single quarter of stagnation.
This is why economists also look at the annual change in economic activity compared with the same quarter of last year. That comparison is useful for a company reviewing its orders and sales against the previous year, although its own results may differ from the economic average.
A technical recession is a signal to look further. Production, consumption, income, credit and confidence offer different views of the economy; none alone explains what will happen to a particular company.
Did Romania enter recession?
Yes. According to the quarterly GDP series published by Eurostat, adjusted for seasonal and calendar effects, Romania’s economy fell by 1.9% in Q4 2025 compared with Q3 2025 and by 0.1% in Q1 2026 compared with Q4 2025. That makes 2 quarters in a row of decline, so the technical threshold of recession is met.
How it was calculated, and the source
- Source · Eurostat, dataset namq_10_gdp, Romania, real GDP, seasonally and calendar adjusted.
- Calculation · each bar is the change versus the previous quarter as published by Eurostat; we do not calculate it. Latest quarters: Q3 25 0.0% · Q4 25 −1.9% · Q1 26 −0.1% · Q2 26 0.0%.
- Technical recession · 2 quarters in a row with the bar below zero.
The chart highlights the fall in Q4 2025, the largest in the period shown. It is followed by a small decline in Q1 2026 and stagnation in Q2 2026. Comparing these with earlier quarters shows how much the economy’s growth rate changed at the end of 2025.
In Q2 2026, the change from the previous quarter was 0%, so the quarterly decline stopped without recovering the earlier loss. For a company, this is a reason to check its own order book against its sector, rather than assume national stagnation describes every customer.
In February, INS published early GDP estimates that met the commonly used technical-recession threshold. The chart uses the Eurostat figures available when this article was prepared. GDP estimates can be revised, so the dates and version of the series matter when comparing reports.
How it was calculated, and the source
- Source · Eurostat, dataset namq_10_gdp, Romania, real change versus the same quarter last year, seasonally and calendar adjusted.
- Calculation · The figure is published directly by the source as an annual change (−2.0%); we do not calculate it.
- Rank · The last time the annual change was as weak or weaker was in Q4 2020 (−2.5), 22 quarters ago.
- Trend · The annual change was negative in every quarter from Q4 2025 to Q2 2026, that is 3 in a row.
The second chart shows the same economy, measured against last year. Compared with Q2 2025, GDP is 2.0% lower. This is the 3rd negative annual change in a row and the weakest since Q4 2020, the winter of the pandemic. The last annual increase was in Q3 2025, at +1.4%.
The two charts answer different questions. The quarterly fall stopped in Q2 2026, but GDP was still below Q2 2025. The year-on-year figure does not tell us when an individual company will feel a change in demand.
What happens in a recession?
Production, consumption and confidence do not always move together. In the 2026 data, industrial output, retail volume and real wages were all below their year-earlier levels. How that affects a company’s sales depends on its customers and the goods it sells.
Industry: 7 months in a row below last year
Industry produced 6.3% less in July than a year earlier, according to INS data, the 7th consecutive month of decline. Manufacturing, meaning the factories themselves, fell by 7.2%, according to Eurostat. On the calendar-adjusted Eurostat series, industry as a whole gives the same figure, −6.3%, so the decline is not an effect of working days.
How it was calculated, and the source
- Source · INS, TEMPO Online database, matrix IND104N, index 2021=100, unadjusted series; search for matrix IND104N in TEMPO.
- Calculation · Calculated by us, because the source publishes only the level: (94.5 / 100.9 − 1) × 100 = −6.34%, that is July 2026 versus July 2025.
- Trend · The annual change was negative in every month from January 2026 to July 2026, that is 7 in a row.
The chart shows industrial production alternating between annual growth and decline in 2025, followed by consistently negative annual changes in 2026.
The bars from January to July are all red, and the lowest is May, at −7.4%. The series is raw, meaning it is not seasonally adjusted, which is why we read it only against the same month of last year.
Lower industrial output can mean weaker demand for some suppliers, but this series does not forecast their orders. A company selling to factories should compare the sector data with its customers’ order plans and its own pipeline.
Not all industries fall by the same amount.
How it was calculated, and the source
- Source · Eurostat, dataset sts_inpr_m, Romania, calendar adjusted series.
- Calculation · each bar is the annual change published by the source for its branch; we do not calculate it (Jul 26 versus the same month last year).
- Branches · Food industry (C10) · Beverage manufacturing (C11) · Chemical industry (C20) · Rubber and plastics (C22) · Metal products (C25) · Machinery and equipment (C28) · Auto industry (C29) · Furniture manufacturing (C31).
Chemicals fell by 14.8%, furniture by 11.7%, the 8th month of decline, and the car industry by 10.1%, the 10th month of decline. Machinery and equipment fell by 9.6%, and metal products by 9.1%. The size of the decline differs sharply by branch.
Food production was close to its year-earlier level, at −0.7%, while plastics was the only tracked branch with growth, at +2.1%. These figures show a difference between branches; they do not establish what caused it.
For a manufacturer, the relevant comparison is between its own sales and its sector’s performance. If both measure the same market using comparable indicators, a smaller decline in the company’s sales may indicate a gain in market share, while growth below the market rate may indicate a loss.
Consumption: 12 months below last year’s level
Consumption moves in the same direction. Retail trade fell by 6.2% in July, according to INS, and sales have been below last year’s level for 12 months in a row.
How it was calculated, and the source
- Source · INS, TEMPO Online database, matrix COM1071, index 2021=100, volume, unadjusted series; search for matrix COM1071 in TEMPO.
- Calculation · Calculated by us, because the source publishes only the level: (118.1 / 125.9 − 1) × 100 = −6.20%, that is July 2026 versus July 2025.
- Trend · The annual change was negative in every month from August 2025 to July 2026, that is 12 in a row.
The chart shows the moment consumption changed direction. Until July 2025 the bars were positive, and from August 2025 they went below zero and have not returned. The weakest month on the chart was January 2026, at −8.3%.
The series measures the volume of goods sold after adjusting for price changes. It does not show that every household bought fewer items or paid more for each product. The wage data below add context on purchasing power.
Eurostat measures the same trade on a calendar-adjusted series and reaches −5.7% in July, while trade in food, beverages and tobacco fell by 5.4%. Basic purchases are falling too, not only those that can be postponed.
How it was calculated, and the source
- Retail trade, volume (G47). Source · Eurostat, dataset sts_trtu_m, Romania, NACE activity G47, sales volume (excluding price effects), index, year 2021 = 100, calendar adjusted (excluding the effect of working days), not seasonally adjusted.
- Calculation · The annual change is the one published by the source: −5.7% versus July 2025. Check on the index: (117.3 / 124.4 − 1) × 100 = −5.71% (differences of one decimal come from the rounding of the published indices).
- Food retail trade, volume. Source · Eurostat, dataset sts_trtu_m, Romania, NACE activity G47_FOOD, sales volume (excluding price effects), index, year 2021 = 100, calendar adjusted (excluding the effect of working days), not seasonally adjusted.
- Calculation · The annual change is the one published by the source: −5.4% versus July 2025. Check on the index: (107.2 / 113.3 − 1) × 100 = −5.38% (differences of one decimal come from the rounding of the published indices).
The two lines stayed below zero for 12 months. Food sales falling alongside total retail volume is a useful signal, but the chart alone cannot tell us how much of the change comes from real income rather than other factors.
Purchasing power: wages grow more slowly than prices
The average net wage reached 5,820 lei (RON) in July, 5.5% above July 2025. Prices, however, rose by 8.2% over the same period, according to the INS consumer price index.
How it was calculated, and the source
- Average net earnings (INS). Source · INS, TEMPO Online database, matrix FOM106G, lei; search for matrix FOM106G in TEMPO.
- Calculation · Calculated by us, because the source publishes only the level: (5820.0 / 5517.0 − 1) × 100 = 5.49%, that is July 2026 versus July 2025.
- Consumer price index (INS), total. Source · INS, TEMPO Online database, matrix IPC102E, % versus the same month of the previous year (from the index with last year = 100 as base); search for matrix IPC102E in TEMPO.
- Calculation · INS publishes the index versus the same month last year = 100. Annual inflation = index − 100 = 8.16%.
Until June 2025, the cream line of wages sat above the red line of prices, so real wages were rising. Since July 2025, prices have risen faster than wages. In May 2026, inflation reached 10.8%, while wage growth fell to 3.2%.
Nominal wages rise, but the amount received buys fewer goods and services than a year ago. We calculated the change in real wages from the two INS series by adjusting wage growth for price growth, using (1 + 5.49%) / (1 + 8.16%) − 1 = −2.47%.
How it was calculated, and the source
- Source · UNRIVALS calculation from two INS series (net earnings and inflation), the real growth of net earnings; it is not published as such by any institution.
- Calculation · Net earnings (INS, FOM106G): 5820 lei in July 2026 versus 5517 lei in July 2025, that is a nominal increase of 5.49%. Inflation (INS, IPC102E, index versus the same month last year = 108.16): 8.16%. Real growth = (1 + 5.49%) / (1 + 8.16%) − 1 = −2.47%.
- Rank · The last time the annual change was as good or better was in July 2025 (−2.4), 12 months ago.
- Trend · The annual change was negative in every month from July 2025 to July 2026, that is 13 in a row.
Rounded, the average real net wage was 2.5% below its year-earlier level, the 13th consecutive month of decline. The deepest point was May 2026, at −6.9%. July’s smaller decline reflects slower price growth and faster nominal wage growth in these series.
July’s decline was smaller, but real wages were still below their year-earlier level. Future readings depend on both wages and prices. Consumer-facing companies can follow this series alongside their own sales; it does not explain retail volume on its own.
Inflation in Romania compared with the rest of the European Union
How it was calculated, and the source
- Annual inflation (HICP), Romania. Source · Eurostat, dataset prc_hicp_minr, Romania, consumption group TOTAL, annual inflation rate.
- Calculation · The figure is published directly by the source as an annual change (6.3%); we do not calculate it.
- Annual inflation (HICP), EU average. Source · Eurostat, dataset prc_hicp_minr, area EU27_2020, consumption group TOTAL, annual inflation rate.
- Calculation · The figure is published directly by the source as an annual change (3.2%); we do not calculate it.
The Harmonised Index of Consumer Prices (HICP) is the official consumer inflation measure used across the European Union and the euro area.
Inflation measured by the Union’s common methodology was 6.3% in August in Romania, against 3.2% for the EU average, according to Eurostat. The chart shows how far Romania has moved from the rest of the Union over the last year, peaking in May 2026 and easing only from June.
Inflation of 6.3% is the lowest since June 2025, according to the Eurostat series. This index measures consumer prices, not each company’s costs. An exporter should read the gap with the EU average alongside its own input costs and prices in the markets it serves.
What is the current state of Romania’s economy?
At the end of September 2026, Romania’s economy is in stagnation after a technical recession. GDP is no longer falling from one quarter to the next, yet production, consumption and real income are all below last year’s level, and the leu is at its weakest against the euro in the series we track.
The table below puts the main indicators side by side. Each comes directly from the database of the institution that publishes it.
| Indicator | Latest period | Compared with last year | Source |
|---|---|---|---|
| Real GDP | Q2 2026 | −2.0% | Eurostat |
| Industrial production | July 2026 | −6.3% | INS |
| Retail trade, volume | July 2026 | −6.2% | INS |
| Real net wage | July 2026 | −2.5% | UNRIVALS calculation, INS data |
| Wholesale on a fee or contract basis (trade intermediaries), volume | July 2026 | −24.3% | Eurostat |
| Passenger car sales, volume | July 2026 | −16.2% | Eurostat |
| Newly registered companies | August 2026 | −29.2% | National Trade Register Office (ONRC) |
| Construction works | July 2026 | +1.8% | Eurostat |
The declines in trade intermediation and car sales call for a closer look. We examine each in its own chart.
B2B distribution: the chain contracts faster than consumption
How it was calculated, and the source
- Source · Eurostat, dataset sts_trtu_m, Romania, calendar adjusted series.
- Calculation · each bar is the annual change published by the source for its branch; we do not calculate it (Jul 26 versus the same month last year).
- Branches · Wholesale trade (G46) · Trade intermediaries (G461) · Wholesale of food and beverages (G463) · Wholesale of household goods (G464) · Other specialised wholesale (G467).
Wholesale trade fell by 5.2% in total, its biggest annual decline since December 2024. The average hides large differences, however.
Trade intermediaries fell by 24.3% and wholesale of food by 6.8%, the 16th month of decline. Wholesale of household and cleaning products, meanwhile, grew by 9.9%, the 11th month of growth.
Specialised wholesale, which includes fuels, hardware, installations and chemicals, rose by 1.3%, its 6th month of growth. The branch you sell in matters more than the wholesale average. Household products and food followed different paths; each distributor’s results need to be checked separately.
How it was calculated, and the source
- Source · Eurostat, dataset sts_trtu_m, Romania, NACE activity G461, sales volume (excluding price effects), index, year 2021 = 100, calendar adjusted (excluding the effect of working days), not seasonally adjusted.
- Calculation · The annual change is the one published by the source: −24.3% versus July 2025. Check on the index: (106.8 / 141.1 − 1) × 100 = −24.31% (differences of one decimal come from the rounding of the published indices).
- Rank · The last time the annual change was as weak or weaker was in August 2019 (−27.7), 83 months ago.
- Trend · The annual change was negative in every month from April 2026 to July 2026, that is 4 in a row.
This series covers wholesale trade on a fee or contract basis. It recorded the largest decline among the series examined in the article. In December 2025, volume was 37.9% above its year-earlier level. From April 2026, the annual change turned negative and fell month by month to −24.3% in July, the biggest annual decline since August 2019.
The series shows a sharp fall in this branch, but it does not show suppliers’ stock levels or future orders. A company selling through agents or distributors has a reason to ask those partners about orders and stock plans for winter.
The car trade has its biggest decline since May 2020
How it was calculated, and the source
- Source · Eurostat, dataset sts_trtu_m, Romania, NACE activity G451, sales volume (excluding price effects), index, year 2021 = 100, calendar adjusted (excluding the effect of working days), not seasonally adjusted.
- Calculation · The annual change is the one published by the source: −16.2% versus July 2025. Check on the index: (135.7 / 162.0 − 1) × 100 = −16.23% (differences of one decimal come from the rounding of the published indices).
- Rank · The last time the annual change was as weak or weaker was in May 2020 (−43.7), 74 months ago.
- Trend · The annual change was negative in every month from June 2026 to July 2026, that is 2 in a row.
Passenger car sales fell by 16.2%, the biggest annual decline since May 2020, the months of the pandemic. The chart shows a market that stayed mostly positive until May 2026, with a few weak months, and then fell in June and July.
The car trade as a whole, including service and parts, fell by 12.9%, also the biggest annual decline since May 2020.
A car purchase can be postponed when household budgets tighten or financing becomes harder to afford. For a dealer, the market decline is a reason to examine enquiries, lost offers and why customers choose to buy or wait.
Construction: works grow, permits fall
How it was calculated, and the source
- Construction output (F). Source · Eurostat, dataset sts_copr_m, Romania, NACE activity F, production volume, index, year 2021 = 100, calendar adjusted (excluding the effect of working days), not seasonally adjusted.
- Calculation · The annual change is the one published by the source: 1.8% versus July 2025. Check on the index: (175.5 / 172.4 − 1) × 100 = 1.80% (differences of one decimal come from the rounding of the published indices).
- Building permits for dwellings (INS). Source · INS, TEMPO Online database, matrix LOC108C, building permits for dwellings, number; search for matrix LOC108C in TEMPO.
- Calculation · Calculated by us, because the source publishes only the level: (6574.0 / 8812.0 − 1) × 100 = −25.40%, that is July 2026 versus July 2025.
Construction is growing, at +1.8% in July, the 6th consecutive month of growth. Building permits for housing, however, fell by 25.4% in July, according to INS, after rising by 43.0% in April.
The change in permits fluctuates considerably from month to month, so any single month needs to be read in the context of the series. The INS construction series shows a similar trend to Eurostat’s, with +2.0% in July.
Permits concern projects that may become building sites later, so their decline may signal a future slowdown even while current construction activity grows. The series measure different stages; one month cannot tell us how many projects will be built.
Money and confidence: what companies feel
Exchange rates, borrowing costs and confidence describe different risks. A company earning in lei and paying in euros needs to check its currency exposure; one that borrows needs to examine the interest terms in its loan agreement.
How it was calculated, and the source
- Source · European Central Bank, series EXR.D.RON.EUR.SP00.A, lei per 1 euro.
- Calculation · Calculated by us: 5.2788 lei (30 September 2026) versus the rate of 30 September 2025, that is 3.90%.
- Rank · The value is the highest in the entire available series, which starts in 2 January 2015.
On 30 September, the euro reached 5.2788 lei, the highest rate in the ECB series, which starts in 2015, 3.9% above the level of a year ago.
The official rate of the National Bank of Romania (BNR) on the same day is 5.2785 lei, practically identical. Compared with the same date a year earlier, a fixed euro payment requires more lei at this rate. The effect on a company earning in lei depends on timing, currency hedges and pricing.
The BNR policy interest rate is 6.5%, and the 3-month ROBOR, used to price some lei loans, is 5.98%. Those reference rates do not show what every company pays: the actual cost depends on the contract, the bank’s margin and when the rate resets.
Many founders tell us their costs have risen faster than they have been able to raise prices, so the difference has come out of their margin. A stronger euro or an interest rate that stays high is one more line in a calculation that no longer adds up.
How it was calculated, and the source
- Source · Eurostat, dataset ei_bssi_m_r2, Romania, indicator BS-ESI-I, seasonally adjusted.
- Calculation · Calculated by us, because the source publishes only the level: (92.2 / 94.1 − 1) × 100 = −2.02%, that is September 2026 versus September 2025.
- Trend · The annual change was negative in every month from January 2025 to September 2026, that is 21 in a row.
The economic sentiment indicator that Eurostat calculates for Romania is 92.2 in September, below the threshold of 100 that marks the historical average, and below the level of a year ago for the 21st month in a row. The chart shows an almost flat line for over a year, not rising towards the threshold of 100.
The indicator has stayed below its historical average. It gathers responses from companies in industry, services, trade and construction, and from consumers. It does not show whether every company is postponing investment; that needs to be checked against decisions and data in each sector.
How it was calculated, and the source
- Source · Eurostat, dataset ei_bsco_m, Romania, calendar adjusted series.
- Calculation · annual change, calculated from the published indices (Sep 26 versus the same month last year).
- Branches · Consumer confidence (ei_bsco_m) · Industry confidence (ei_bsin_m_r2) · Services confidence (ei_bsse_m_r2) · Retail trade confidence (ei_bsrt_m_r2) · Construction confidence (ei_bsbu_m_r2).
The chart shows confidence by sector. Consumers are 1 point below the level of a year ago, industry 1.4 points and services 2.8 points. Retail trade and construction are the only sectors more optimistic than a year ago, at +8.4 and +1.8 points.
One possible explanation is that falling inflation supports retailers’ expectations while service providers face reduced customer budgets. The confidence data alone do not establish that explanation. Consumer confidence remains very low, at a balance of −32.5 points, so retailers’ optimism still needs to be checked against sales performance.
Not everything is falling: where is the market growing?
The market does not fall uniformly, and the exceptions are as useful as the average. Eurostat data for July show several sectors that are growing right now. For the companies in them, or those that sell to them, the recession looks different from the press headlines.
How it was calculated, and the source
- Source · Eurostat, dataset sts_setu_m, Romania, NACE activity J62, turnover in value (prices included), index, year 2021 = 100, calendar adjusted (excluding the effect of working days), not seasonally adjusted.
- Calculation · The annual change is the one published by the source: 5.0% versus July 2025. Check on the index: (219.9 / 209.4 − 1) × 100 = 5.01% (differences of one decimal come from the rounding of the published indices).
- Trend · The annual change was positive in every month from December 2024 to July 2026, that is 20 in a row.
IT services turnover rose by 5.0% in value, the 20th month of growth. The annual rate is below December 2025’s 25.4%. This series includes price effects, so it cannot by itself tell us how service volumes changed.
How it was calculated, and the source
- Services to businesses (INS), volume. Source · INS, TEMPO Online database, matrix PSC108P, index 2021=100, volume (production indices), unadjusted series; search for matrix PSC108P in TEMPO.
- Calculation · Calculated by us, because the source publishes only the level: (133.5 / 140.6 − 1) × 100 = −5.05%, that is July 2026 versus July 2025.
- Turnover of market services (H-N excl. K). Source · Eurostat, dataset sts_setu_m, Romania, NACE activity H-N_X_K, turnover in value (prices included), index, year 2021 = 100, calendar adjusted (excluding the effect of working days), not seasonally adjusted.
- Calculation · The annual change is the one published by the source: 1.4% versus July 2025. Check on the index: (182.5 / 180.0 − 1) × 100 = 1.39% (differences of one decimal come from the rounding of the published indices).
For business services, INS reports −5.1% in volume, while the Eurostat series shows +1.4% in value. The percentages measure different things, and the series also use different adjustments, described below the chart. Prices may contribute to the difference, but the comparison alone does not establish that every customer pays more and orders less often. A provider needs to check invoice values and the volume of services sold separately.
Wholesale of household and cleaning products, up 9.9% in volume, is the second example. This branch grew while wholesale trade overall declined. That does not mean every distributor in the branch performed better.
The average can hide large differences between branches. Before changing its budget, a company needs to compare sector figures with its own orders and margins.
UNRIVALS methodology: how did we build a macro report from official sources?
For this analysis, we organised a catalogue of 51 series from sources including Eurostat, INS, BNR, the European Central Bank, ONRC, ACAROM and CNSP. The workflow is designed for weekly source checks, but each institution publishes on its own schedule. The changes and historical comparisons used here were calculated from the series available when the article was prepared.
Most figures were taken from the institutions’ databases in structured form. When the source is a release, we keep a link to the passage behind the figure. Readers can then check the data and calculation supporting a conclusion.
Some figures, such as the BNR balance of payments, appear in press releases. For those, we compared the values used here with the original release.
That is how we found, for example, in the BNR release of September, that the current account deficit reached 16.3 billion euro between January and July 2026.
Claims about records and consecutive months need to be checked against the relevant full series. A figure in this article should therefore be read with its date and source, rather than as a forecast for a reader’s company.
This analysis offers a starting point for discussing a client’s market. Sector data need to be considered alongside its orders, margins and sales team’s observations before they support a decision.
What does a company do in a contracting market?
In a contracting market, companies need to understand which purchases customers postpone and what they continue to buy. A clear reason to choose your company helps buyers understand why your product or service deserves to remain in their budget.
At UNRIVALS, cognitive ownership describes a buyer remembering a company when a need in its category arises. It is an aim built through positioning and customer experience, not a promise that people will buy automatically or that distribution and advertising are unnecessary.
We wrote at length about cognitive ownership and how it is built.
Recession makes this principle more urgent because a buyer with a limited budget needs reasons to trust a supplier before choosing. Recognising the company’s name and understanding its offer can help it enter the buyer’s shortlist.
The second principle concerns the budget. Investment in strategy and an offer the company can deliver should be assessed alongside channel optimisation. Clearer positioning may help lower customer acquisition cost, but the effect must be measured rather than assumed.
In a recession, a company may also cut spending that supports differentiation if it judges that work solely by the contracts it generates immediately.
If all remaining spending goes to the channel that worked before, the company risks depending on shrinking demand. Optimising that channel can improve results, but it needs to be accompanied by a review of the offer, the audience and the reasons customers buy from the company.
We wrote separately about how a budget is split, in the guide to the marketing budget.
The third principle concerns the clarity of the offer. If buyers must work hard to understand what they get and whom a service suits, the message may make their decision harder. We call that difficulty cognitive cost. Clarity can help, but its effect on enquiries and acquisition costs must be checked in the data.
The full mechanism is described in the article on what an unclear brand costs.
When buyers have tighter budgets, a message needs to give them clear criteria for choosing. If they cannot see the difference between offers, they may compare mostly on price. Enquiry data and sales conversations can show whether that is happening for a particular company.
We often talk to founders who cut their price because it was the only lever within reach, and the volume did not follow. Their fear is losing the order and the margin at the same time.
On which layers is a recession felt in a company?
The four layers below offer a way to check where a company may feel pressure. A recession does not affect them in the same order for everyone.
We use the same four layers in our audits, covering L1, performance marketing; L2, revenue and the commercial process; L3, AI Brain; and L4, positioning and category. The analysis connects advertising and sales results with market information and the company’s positioning.
- L4 · Positioning and category Is the reason to choose the offer clear?
- L3 · AI Brain, orchestration Does market data reach decisions in time?
- L2 · Revenue and commercial process What is changing in orders and negotiations?
- L1 · Performance marketing How are enquiries and their cost changing?
The figure can be read from the bottom up, starting with results you can observe. At L1, compare enquiries and their cost with similar periods. If results change, examine the offer and enquiry quality too; advertising figures alone cannot identify the cause.
At L2, track orders, sales-cycle length and reasons for lost offers. If negotiations often focus solely on price, the team should check whether it explains the offer’s value and limits clearly. We cover this layer in the guide to marketing strategy.
At L3, which we call AI Brain, check whether relevant market data reach decision-makers and are compared with the company’s own data. Seeing a sector figure quickly helps only if the team knows what question to ask next.
At L4, ask whom the offer suits and what verifiable reason the customer has to choose it. The answer helps sales and communication, but the customer’s decision also depends on budget, availability and experience.
What do the figures mean for production and distribution?
For companies in production and B2B distribution, the 2026 figures show volume declines in several branches, but they do not tell us how every buyer negotiates. The differences are large, from −24.3% for trade intermediaries to +9.9% for wholesale of household products. Each company needs to compare the relevant branch with its own sales.
In food production, factories stayed close to last year’s level, at −0.7%, while food retail fell and food wholesale has been declining for 16 months. For a producer selling through distributors, these figures are a reason to check both distributor orders and consumer demand before deciding how to expand sales.
In B2B distribution, the fall in trade intermediation is a reason to check orders and stock levels in the channel. When several distributors sell the same brands, service, specialisation and delivery time can give buyers criteria beyond price. The company needs to find out which of these actually matter to its customers.
For suppliers of machinery, parts or components, the machinery, metal and automotive sectors are all declining faster than industry overall. An industrial buyer may postpone investment, so the supplier needs to find out which projects are continuing and which have stopped. Maintaining the relationship and trust can help it remain on the shortlist when the customer resumes purchasing.
In construction, works are growing while permits are falling. A supplier of installations or materials can follow both series, then ask customers which sites are active and which new projects have approval. Permits do not automatically turn into orders.
What founders tell us about the recession
The figures above describe the market. Conversations with founders describe it from the inside, and three things come up more often than the rest.
The first is the quiet in the order book. Founders describe it without drama: “Orders are coming in less often and smaller, and clients have put some projects on hold.” Under that sentence you can hear the worry of someone who has a team and capacity and does not know how long they can keep both full.
The second is cost pressure. “Energy and salaries have gone up, but the customer won’t accept price rises, so I pay the difference.” What sits underneath is the fatigue of working as hard as before and watching the margin thin out.
The third is price. Some founders admit they cut it because they had no other quick move. “If we cut again we earn nothing, and if we don’t, we lose the order.” Behind that sentence is the fear of being caught between two risks with no third option.
If you recognise one of these situations, the article gives you the figures for your sector, and the exercise that follows helps you find what reason to choose a customer with a tightening budget still has.
The exercise for your company
This 30-minute exercise with the sales team helps identify risks worth investigating. Answers to three questions cannot by themselves measure the company’s exposure.
The first question concerns your industry and the companies buying products or services from you. In the charts above, or in Eurostat and INS data, look for both your sector and your customers’ sectors. If their activity is falling, ask which orders they might postpone; the sector indicator signals a risk without telling you exactly when it will affect your sales.
The second question is about the reason to choose you. If a customer had to choose tomorrow between your company and the lowest-priced offer on the market, what reason would they have to stay with you, besides price? Write the reason in a single sentence, then ask yourself whether a competitor could use the same sentence.
The third question is about budget. Look at where marketing spending went in the last six months and what results came from short-term activities versus work that builds a position, content or relationships. Compare their costs and effects before moving the budget.
If the first two answers gave you pause, you can request a diagnostic as a starting point. Your answers can show what to investigate; to learn how customers see the company, you need to ask them.
What do the official releases say?
Besides the databases, we consulted releases from the BNR, the Association of Automobile Manufacturers of Romania (ACAROM) and the National Commission for Strategy and Forecasting (CNSP). Read the figures below with the period and definition used by each institution.
External money: the deficit grows, foreign investment falls
The current account deficit reached 16.3 billion euro between January and July 2026, against 15.7 billion euro in the same period of 2025, according to the BNR. The services surplus was 680 million euro larger. The current account also includes goods, income and transfers; one component cannot explain the change in the deficit on its own.
The net flow of foreign direct investment was smaller. For January to July 2026, the BNR reported 1.1 billion euro, against 4.9 billion euro in the same period of 2025. This comparison cannot by itself explain why some industrial branches declined; investment projects and production run on different timelines.
Total external debt reached 234.9 billion euro at the end of July, 6.4 billion euro more than at the start of the year. A borrowing company needs to check its actual costs in the loan terms; changes in external debt alone do not show how the interest paid by each borrower changes.
The car market: the hardest month of the year
Registrations of new passenger cars fell in August by 36.85% compared with August 2025, according to ACAROM, to 9,552 cars. Between January and August, 86,142 new cars were registered, 10.12% fewer than last year.
The fall does not hit all engine types equally. Fully electric cars grew by 50%, to 7,018, and hybrids by 2.1%, to 49,282, while petrol fell by 31.3% and diesel by 39.7%. The car market is changing its composition while it falls.
Second-hand cars registered for the first time in Romania also fell, by 13.4%, to 199,570. Dacia remains the top brand, with 17,114 new cars. New-car registrations in the European Union grew by 5.3% over the same period, while Romania recorded a decline. The comparison shows different trends without establishing their causes.
The official forecast: 2026 close to zero, 2027 better
In its spring forecast of May 2026, CNSP estimates GDP growth of only 0.1% in 2026 and 2.2% in 2027. For inflation, the forecast gives an annual average of 7.9% in 2026 and 5.3% at the end of the year.
CNSP forecasts a 1.9% fall in real wage earnings in 2026. It estimates the current account deficit at 26.8 billion euro for the full year. This is a scenario published in May, not an observed result or a set of assumptions a company must automatically adopt in its budget.
Other signals in the data
Not all the figures below made it into the article’s charts, but they are worth tracking for any company planning its next quarter. Each figure has its source and calculation in the appendix at the end.
- New companies. 8,090 registrations in August, 29.2% fewer than in August 2025, the biggest annual decline since August 2024, according to the National Trade Register Office (ONRC). Deregistrations fell as well, by 19.0%.
- Industry, by branch. In July, according to Eurostat, chemicals fell by 14.8%, furniture by 11.7%, the 8th month of decline, the car industry by 10.1%, the 10th month of decline, machinery and equipment by 9.6%, and metal products by 9.1%. Of the tracked branches, the only one in positive territory is plastics, at +2.1%.
- Distribution, by branch. Wholesale of household and cleaning products grew by 9.9% in volume, the 11th month of growth. Specialised wholesale, with fuels, hardware, installations and chemicals, rose by 1.3%, and wholesale of food fell by 6.8%, the 16th month of decline.
- IT services. Turnover grew by 5.0%, in value, with prices included, the 20th month of growth.
- Confidence. In retail trade, September confidence is +8.4 points above the level of a year ago, the best change since April 2024, according to Eurostat. In services, confidence has been below the level of a year ago for the 21st month in a row.
- Money. The BNR interest rate is 6.5%, the 3-month ROBOR is 5.98%, and the ECB deposit rate is 2.5%. Food inflation came down to 2.1%, the lowest since July 2024, while total inflation is 6.3%, against 3.2% for the EU average, according to Eurostat.
- Labour. Unemployment is 6.4% in July, +0.5 points above the level of a year ago. The average net wage reached 5,820 lei, 5.5% above July 2025, yet below inflation.
- Official forecast. The National Commission for Strategy and Forecasting estimates GDP growth of +0.1% for 2026, in its spring forecast of May 2026.
Which indicators do we track, and where are they now?
The text selects a few figures for discussion; the appendix shows the series used to prepare it. The table groups values available when the analysis was prepared, with their period, change or trend, and source.
The table was generated from data processed for this article; it is not a live feed. For some series, such as interest rates or the daily exchange rate, the comparison uses the same day or month of the previous year.
Inflation
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Annual inflation (HICP), Romania % versus the same month last year | August 2026 | 6.3 | 6.3% | weakest since June 2025 | Eurostat |
| Food inflation (HICP CP01) % y/y | August 2026 | 2.1 | 2.1% | weakest since July 2024 | Eurostat |
| Annual inflation (HICP), EU average % y/y | August 2026 | 3.2 | 3.2% | – | Eurostat |
| Consumer price index (INS), total % versus the same month of the previous year (from the index with last year = 100 as base) | July 2026 | 8.2 | 8.2% | weakest since July 2025 | INS |
Industry
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Industrial production, total (B-D) index 2021=100, calendar adjusted | July 2026 | 92.8 | −6.3% | weakest since March 2025 · 7 months of decline | Eurostat |
| Manufacturing (C) index 2021=100, CA | July 2026 | 94.8 | −7.2% | weakest since March 2025 · 7 months of decline | Eurostat |
| Food industry (C10) index 2021=100, CA | July 2026 | 117.4 | −0.7% | – | Eurostat |
| Beverage manufacturing (C11) index 2021=100, CA | July 2026 | 117.3 | −5.4% | 3 months of decline | Eurostat |
| Chemical industry (C20) index 2021=100, CA | July 2026 | 71.2 | −14.8% | 3 months of decline | Eurostat |
| Rubber and plastics (C22) index 2021=100, CA | July 2026 | 96.6 | 2.1% | – | Eurostat |
| Metal products (C25) index 2021=100, CA | July 2026 | 105.3 | −9.1% | 3 months of decline | Eurostat |
| Machinery and equipment (C28) index 2021=100, CA | July 2026 | 93.8 | −9.6% | weakest since April 2025 | Eurostat |
| Auto industry (C29) index 2021=100, CA | July 2026 | 89.9 | −10.1% | 10 months of decline | Eurostat |
| Furniture manufacturing (C31) index 2021=100, CA | July 2026 | 90.4 | −11.7% | 8 months of decline | Eurostat |
| Industrial production (INS), unadjusted series index 2021=100, unadjusted series | July 2026 | 94.5 | −6.3% | 7 months of decline | INS |
Consumption
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Retail trade, volume (G47) index 2021=100, volume, CA | July 2026 | 117.3 | −5.7% | 12 months of decline | Eurostat |
| Food retail trade, volume index 2021=100, volume, CA | July 2026 | 107.2 | −5.4% | 12 months of decline | Eurostat |
| Retail trade (INS), volume index 2021=100, volume, unadjusted series | July 2026 | 118.1 | −6.2% | 12 months of decline | INS |
Auto
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Auto trade (G45), volume index 2021=100, volume, CA | July 2026 | 128.1 | −12.9% | weakest since May 2020 | Eurostat |
| Motor vehicle sales (G451), volume index 2021=100, volume, CA | July 2026 | 135.7 | −16.2% | weakest since May 2020 | Eurostat |
| New car registrations (ACAROM) cars | August 2026 | 9,552 | – | – | official release |
Distribution
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Wholesale trade (G46), volume index 2021=100, volume, CA | July 2026 | 105.6 | −5.2% | weakest since December 2024 · 4 months of decline | Eurostat |
| Trade intermediaries (G461), volume index 2021=100, volume, CA | July 2026 | 106.8 | −24.3% | weakest since August 2019 · 4 months of decline | Eurostat |
| Wholesale of food and beverages (G463), volume index 2021=100, volume, CA | July 2026 | 87.1 | −6.8% | 16 months of decline | Eurostat |
| Wholesale of household goods (G464), volume index 2021=100, volume, CA | July 2026 | 112.9 | 9.9% | 11 months of growth | Eurostat |
| Other specialised wholesale (G467), volume index 2021=100, volume, CA | July 2026 | 93.7 | 1.3% | 6 months of growth | Eurostat |
Services
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Turnover of market services (H-N excl. K) index 2021=100, nominal value, CA | July 2026 | 182.5 | 1.4% | 5 months of growth | Eurostat |
| IT services (J62), turnover index 2021=100, value, CA | July 2026 | 219.9 | 5.0% | 20 months of growth | Eurostat |
| Services to businesses (INS), volume index 2021=100, volume (production indices), unadjusted series | July 2026 | 133.5 | −5.0% | weakest since May 2025 | INS |
Construction
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Construction output (F) index 2021=100, CA | July 2026 | 175.5 | 1.8% | 6 months of growth | Eurostat |
| Construction works (INS) index 2021=100, unadjusted series | July 2026 | 177.7 | 2.0% | 6 months of growth | INS |
| Building permits for dwellings (INS) building permits for dwellings, number | July 2026 | 6,574 | −25.4% | weakest since April 2025 | INS |
Confidence
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Consumer confidence balance, points | September 2026 | −32.5 | −1.0 pp | – | Eurostat |
| Economic sentiment indicator (ESI) index, long-term average = 100 | September 2026 | 92.2 | −2.0% | 21 months of decline | Eurostat |
| Industry confidence balance, points | September 2026 | −3.0 | −1.4 pp | 7 months of decline | Eurostat |
| Services confidence balance, points | September 2026 | −3.0 | −2.8 pp | 21 months of decline | Eurostat |
| Retail trade confidence balance, points | September 2026 | 2.7 | 8.4 pp | best since April 2024 · 3 months of growth | Eurostat |
| Construction confidence balance, points | September 2026 | −9.6 | 1.8 pp | best since April 2024 | Eurostat |
Labour
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Unemployment rate (ILO) % of the labour force | July 2026 | 6.4 | 0.5 pp | 25 months of growth | Eurostat |
| Average net earnings (INS) lei | July 2026 | 5,820 | 5.5% | best since June 2025 · 127 months of growth | INS |
| Purchasing power: real net earnings % y/y, real (net earnings deflated by CPI) | July 2026 | −2.5 | −2.5% | best since July 2025 | UNRIVALS calculation |
GDP
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Real GDP, annual change % versus the same quarter last year | Q2 2026 | −2.0 | −2.0% | weakest since Q4 2020 · 3 quarters of decline | Eurostat |
Money
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| EUR/RON exchange rate (BNR) lei per 1 euro | 30 September 2026 | 5.2785 | 3.9% | – | BNR |
| EUR/RON exchange rate (ECB, backup) lei per 1 euro | 30 September 2026 | 5.2788 | 3.9% | record level of the series | ECB |
| BNR policy interest rate % per year | August 2026 | 6.5 | 0.0 pp | – | BIS |
| ROBOR 3M % per year | 30 September 2026 | 6.0 | −0.5 pp | – | BNR |
| ECB deposit facility rate % per year | 16 September 2026 | 2.5 | 0.5 pp | – | ECB |
Companies
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| New company registrations (ONRC) companies | August 2026 | 8,090 | −29.2% | weakest since August 2024 · 4 months of decline | ONRC |
| Company deregistrations (ONRC) companies | August 2026 | 4,531 | −19.0% | – | ONRC |
External accounts
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| Balance of payments current account (cumulative since the start of the year) bn euro | July 2026 | −16.3 | – | – | official release |
Forecasts
| Indicator | Latest period | Value | Versus last year | Rank and trend | Source |
|---|---|---|---|---|---|
| GDP growth forecast (CNSP) % y/y | December 2026 | 0.1 | 0.1% | – | official release |
Frequently asked questions
What is a technical recession?
A technical recession is the practical rule of 2 consecutive quarters in which real GDP falls compared with the previous quarter. It can be checked against figures, but does not by itself describe the depth or duration of a broader economic recession.
Is Romania still in recession?
According to Eurostat data, Romania’s GDP fell in Q4 2025 and in Q1 2026, then stagnated in Q2 2026, at 0.0% compared with the previous quarter. Technically the fall has stopped, yet the economy is 2.0% below the level of a year ago.
Why is purchasing power falling if wages are rising?
Because prices are rising faster. In July 2026, net earnings grew by 5.5% and prices by 8.2%, so a wage buys about 2.5% less than a year ago.
Which sectors grow in a recession?
According to Eurostat series for July 2026, wholesale of household and cleaning products, specialised wholesale and construction works grew in volume. IT services turnover grew in value, while plastics was the only industrial branch tracked here in positive territory. These series do not show the results of every company.
What should a company do in a recession?
Compare figures for its industry and its customers’ branches with its own orders, learn why buyers choose it, and check what each part of the budget delivers. Spending changes should be based on the company’s results, not macroeconomic indicators alone.
