Employers in construction and hospitality rarely lose a foreign hire at the work permit stage. They lose it earlier, at the visa stage, in a consulate abroad, and often without ever learning why.
The numbers below come from the Ministry of Foreign Affairs of Montenegro, released in October 2026 under the Law on Free Access to Information. As far as we can establish, they have not been published before.
The short version
Across 2024, 2025 and the first half of 2026, Montenegro recorded 13,730 long-stay (D) visa applications filed on an employment basis. The overall refusal rate fell from 62.2% to 45.7% over that period, while the volume of applications nearly tripled.
The overall rate, however, tells an employer almost nothing. Refusal rates by nationality in the first half of 2026 ranged from 3% to 98%.
Where the process actually breaks
For a worker who needs a visa, the sequence runs: long-stay visa (viza za duži boravak, type D) issued by a Montenegrin diplomatic or consular mission abroad, then entry, then the residence and work permit (privremeni boravak i rad) handled domestically.
Most employer attention goes to the second stage, because that is the stage the employer can see and influence. The data shows the attrition is concentrated in the first.
In the first half of 2026, 2,299 employment-based D visa applications were refused. Each one is a worker who never reached the country, and in most cases an employer who had already committed to a start date.
Volume and overall outcome
| Period | Applications | Refused | Refusal rate |
|---|---|---|---|
| 2024 (calendar year) | 3,601 | 2,240 | 62.2% |
| 2025 (calendar year) | 5,095 | 2,521 | 49.5% |
| 2026 (1 January – 30 June) | 5,034 | 2,299 | 45.7% |
Two things are happening at once. Demand is rising steeply, with the first six months of 2026 nearly matching the whole of 2025. And the system is refusing a smaller share of what it receives.
The second trend should not be read as a general relaxation. It is substantially a change in composition: the fastest-growing source countries are ones with low refusal rates.
Refusal rates by nationality
The figures below are refusals as a share of all employment-based D visa applications recorded for that nationality in the period.
| Nationality | 2024 | 2025 | 2026 (H1) |
|---|---|---|---|
| Bangladesh | 97% | 95% | 98% |
| Tunisia | 25% | 33% | 83% |
| Kenya | 61% | 75% | 77% |
| Pakistan | 89% | 81% | 75% |
| Nepal | 77% | 62% | 60% |
| Morocco | 59% | 67% | 58% |
| India | 62% | 50% | 41% |
| Egypt | 65% | 59% | 41% |
| Sri Lanka | 52% | 82% | 37% |
| Azerbaijan | — | — | 27% |
| Cuba | 25% | 42% | 25% |
| China | 10% | 11% | 21% |
| Uzbekistan | 34% | 56% | 18% |
| Indonesia | 38% | 38% | 17% |
| Philippines | 23% | 41% | 16% |
| Kazakhstan | 9% | 6% | 3% |
Three patterns deserve an employer's attention.
Some corridors are closed in practice. Bangladesh has been refused at 95% or above in every period measured. On these numbers a recruitment plan built on Bangladeshi labour carries roughly a 2% chance of clearing the visa stage, irrespective of how well the employer's own file is prepared.
Some corridors close suddenly. Tunisia moved from a 25% refusal rate in 2024 to 83% in the first half of 2026. An employer who recruited successfully from Tunisia in 2024, and assumed the same route would work in 2026, would have lost the season.
Some corridors are reliably open. Kazakhstan, the Philippines, Indonesia, Uzbekistan and China were each refused at under a quarter of applications in the first half of 2026.
What changed in 2026: the visa centres
The 2026 records contain a structural change absent from 2024 and 2025. Applications now appear as received through VFS Global visa centres, in Dhaka, New Delhi, Mumbai, Kolkata, Cochin, Baku, Bishkek, Ankara, Istanbul and Dubai.
This matters operationally, because where an application can be filed determines how quickly it can be filed and how far a worker must travel to file it.
A related feature runs through all three years. In a large number of countries, applications for Montenegrin visas are received by embassies of the Republic of Serbia acting on Montenegro's behalf, among them New Delhi, Cairo, Nairobi, Jakarta, Rabat, Tunis, Havana, Abuja, Pretoria, Baghdad and Tehran. Montenegro's own consular network is limited, and this representation arrangement carries much of the volume.
We have asked the Ministry for the consolidated list of missions and visa centres authorised to receive D visa applications, and the nationalities each one serves. This article will be updated when that list arrives.
Two movements the data explains
Two nationalities move so sharply that they are worth separating from the rest.
Azerbaijan appears in the 2026 records as the single largest source of employment-based applications, at 980 in six months, having been negligible before. Turkey appears at 261 applications in 2025 and only 52 in the first half of 2026.
Both movements coincide with changes in Montenegro's visa regime for those two countries, reported in regional press during 2025 and 2026. The underlying government decisions are published in Službeni list Crne Gore, and the exact dates and conditions should be read there rather than taken from press reporting. What the visa statistics show is the consequence: when a visa requirement is introduced, a flow that was previously invisible to this dataset becomes visible, and when it is lifted, the flow disappears from the dataset without the people disappearing.
That is a caution about reading the table as a measure of migration. It measures visa applications, not arrivals.
Planning a season around this
The practical implication is one of sequencing. The nationality question is not a detail to settle after the commercial terms are agreed; on this data it is the single variable that most strongly predicts whether the workers arrive at all.
Check the corridor before signing. A recruitment agreement that names a source country with a 90% refusal rate commits the employer to an outcome the numbers do not support.
Treat past success as perishable. Tunisia and Sri Lanka both moved by more than 40 percentage points within two years, in opposite directions.
Account for the filing location. With intake partly through visa centres, the question is not only which mission covers a country, but which centre, and whether the worker can reach it.
Keep the employer-side file complete regardless. None of the above substitutes for a properly documented employment relationship, and as of this month that documentation carries more weight than it did.
The compliance side
On 19 October 2026 the amendments to the Law on Foreigners were published in Službeni list Crne Gore no. 154/2026. Among them is a tighter definition of illegal employment (nezakonito zapošljavanje), which now expressly reaches working conditions that endanger health and safety or violate human dignity, not only the status of the paperwork.
For construction and hospitality, the two sectors that depend most heavily on foreign labour, this changes the exposure. It is also a reason to look closely at how any intermediary or partner agency abroad sources and documents workers, because the obligations do not stop at the employer's own gate.
Source and method
All figures are taken from records supplied by the Ministry of Foreign Affairs of Montenegro, document UPI no. 17/1-056/26-198 of 26 October 2026, provided under Article 34(8) of the Law on Free Access to Information (Službeni list CG no. 160/25) in response to a request made on 8 October 2026.
The records cover long-stay (D) visas applied for on the basis of employment only, broken down by applicant nationality and by the mission or visa centre at which the application was filed, for calendar 2024, calendar 2025 and 1 January to 30 June 2026.
A note on method, because it limits what can honestly be said. The source tables classify each application by status, and the set of statuses is not identical across the three years: 2024 uses refused, approved and visa expired; 2025 adds visa issued; 2026 drops visa expired. For that reason this article reports refusals as a share of applications, which is unambiguous in all three years, and does not report approval rates, which would require assumptions about how the remaining statuses relate to one another. We have asked the Ministry to confirm the definitions, and this article will be revised if the position differs from our reading.
Percentages are rounded to the nearest whole number. Period totals are as stated in the Ministry's own tables; small differences between the sum of national rows and the stated totals reflect the source document.
