Nine years of cash-register enforcement: 9,302 suspensions, 2014–2022

Between 2014 and 2022 the State Revenue Committee published periodic «CCM-suspensions» bulletins — Excel files listing taxpayers whose cash-register activity had been suspended for non-compliance. 431 such bulletins were recovered from petekamutner.am via the Internet Archive. 426 of them parsed cleanly into a 9,302-record dataset that, taken end-to-end, is the most complete public picture we have of how cash-register compliance was actually enforced over that period.

This page is what those numbers say. The regulatory arc behind cash-register certification is on a sister page; this one is about who got hit and when.

The data we recovered

• Source directory. petekamutner.am/Shared/Documents/_ts/_ti/CCM_Users_Activity_Suspensions/, organised into year subfolders (2014–2022).
• File format. Each bulletin is a 2003-era .xls (later .xlsx) with the same header: «Տեղեկանք տնտեսվարող սուբյեկտներին» — followed by columns for serial number, name, legal type, TIN, registering tax inspectorate, suspended activity type, and the address of the suspended location.
• Coverage. 431 bulletins recovered (2014: 47 · 2015: 47 · 2016: 34 · 2017: 48 · 2018: 76 · 2019: 52 · 2020: 51 · 2021: 52 · 2022: 24).
• Parsed. 426 of 431 (5 unreadable). Bookend bulletins are registered as primary-source artefacts: 3 April 2014archive and 30 March 2022archive.
• Total records aggregated. 9,302.

Year-by-year volume

The first thing the table shows is a near-9× increase in annual volume from 2014 (263) to 2021 (2,331). The per-bulletin average grew from 5.6 to 74.6 over the same span. Whether that reflects intensifying enforcement activity, or just a switch to fewer bulletins of larger size (2022 shipped only 24 published bulletins, versus 47–76 in earlier years), the dataset alone can't separate. What it does establish unambiguously is that the regime was producing more enforcement output per publication, year over year, through to 2021.

Where the suspensions concentrated

Across all 9,302 records:

• By activity type — 87 % were trade. «Առևտուր» (trade) was the suspended-activity label on 8,104 records. Catering came second at 144, services 87, fuel stations 64. The cash-register enforcement regime was overwhelmingly a small-retail-trade enforcement regime.
• By geography — 57 % in Yerevan. 5,339 records had Yerevan in the address. The next four regions are Gyumri (328), Ararat (314), Armavir (274), and Kotayk (258). Concentration in the capital is in line with population, but not perfectly — Yerevan has roughly a third of Armenia's population, so it's over-represented in suspensions by ~1.5×.
• By legal form — 59 % individual entrepreneurs (ԱՁ). 5,466 ԱՁ records vs 3,038 LLC (ՍՊԸ) records. Other legal forms (CJSC, NGO, SNCO, foreign branches) account for under 100 records combined. The enforcement workload was not corporate — it was small operators registered as individuals.

Why the bulletins stop after 2022

After 2022 the CCM_Users_Activity_Suspensions/ folder on src.am no longer receives new entries. The most likely explanation is that publication moved to a different surface during the taxservice.am portal migration that the broader SRC IT consolidation was producing in 2023–2024 (this is also when G-Solutions SNCO took over IT operations). We haven't yet pinned down where the equivalent enforcement-output publication lives now, or whether it's still public at all.

If you find it: the email at the bottom of this site, please.

What the numbers don't say

A few honest caveats about reading too much into the curve:

• «Suspended» is the disposition, not the reason. The bulletins list outcomes; they don't record what triggered each suspension (failure to issue a fiscal receipt, missing register registration, technical malfunction, fiscal data not transmitted, etc.). The underlying reason mix may have shifted over time.
• «Suspended» doesn't necessarily mean «closed». A taxpayer whose activity was suspended could resume after rectifying the issue. The dataset records suspensions, not durations or recurrences.
• Coverage gaps in 2022. Only 24 bulletins shipped in 2022 (less than half the 51-week pace of 2021). Either enforcement collapsed, or — more likely — publication did, with the actual enforcement continuing internally but no longer surfacing publicly.

So treat the numbers as a faithful record of what was published. The relationship between published-suspensions and underlying compliance-actions is itself a research question we don't yet have a clean answer for.

Sources

• 426 of 431 parsed bulletins from petekamutner.am/.../CCM_Users_Activity_Suspensions/ (2014–2022), recovered via Internet Archive.
• Two bookend bulletins registered as primary-source artefacts (3 April 2014archive · 30 March 2022archive); the full series remains on Internet Archive at the path above.
• Aggregated dataset (records · year · activity type · region · legal form) preserved as a JSON file alongside the source recovery; can be re-run by anyone with xlrd and openpyxl.