Note on figures cited below. Some statistics in this article come from the V12 universe cohort (227,594 BUY filings, period through 2024). Since 2025-01-01 the live out-of-sample BUY universe (28 markets) shows a 52.96 percent win rate and a +8.59 percent mean 90-day return on 20,688 filings. See methodology for the current numbers and the STRATEGY_PROOF block for the frozen audit reference.
A migration can break a dataset more efficiently than a market crash.
Italy is a fine example of a regulatory truth that data people learn the hard way. A disclosure regime can remain legally intact while becoming operationally awkward.
For insider transaction watchers, the Italian story is less about a single rule change than about a publication path changing underneath the rulebook. Historically, users could associate CONSOB with the public face of regulated disclosures. Over time, however, publication and storage shifted toward the Italian system of authorised storage mechanisms, with eMarket STORAGE becoming the main practical destination for many users seeking internal dealing announcements. The result is not doctrinal chaos. It is something more annoying, namely a chain of custody problem for public information.
Our own database offers a blunt illustration.
Zero is a suspiciously tidy number for a major European market. It usually means the pipeline is broken, the source moved, or both.
The migration matters because insider dealing data is only as useful as its retrieval path. If a market’s disclosures are dispersed across regulator pages, storage platforms, and issuer websites, then completeness becomes a methodological assumption rather than a property of the dataset. That is bad for journalism, bad for backtests, and excellent for creating false confidence in comparative charts.
The relevant legal backbone for listed-company insider transactions in Europe is the Market Abuse Regulation, especially Article 19. Persons discharging managerial responsibilities, and persons closely associated with them, must notify transactions once the reporting threshold is reached. Issuers and market participants then need a publication and storage pathway.
In Italy, CONSOB remains the securities regulator and the supervisory authority. But that does not mean every disclosure lives permanently on CONSOB’s own domain in a way that is easy to scrape, search, or archive. Over time, dissemination has been operationalised through the broader Italian system for regulated information, where authorised storage mechanisms and dissemination services play a central role. eMarket STORAGE, operated within the Italian regulated-information infrastructure, became one of the key public repositories users encounter in practice.
That distinction matters. Regulation tells you what must be disclosed. Infrastructure determines where you can still find it six months later.
There are sensible reasons many researchers begin with CONSOB.it:
Italy is less obliging. A researcher arriving at CONSOB may find legal texts, notices, market structure information, and references to regulated information systems, but not a neat, regulator-hosted, longitudinal insider-transactions archive in the style many data teams hope for. The public trail often runs onward to storage mechanisms, issuer disclosures, Borsa Italiana-linked infrastructure, or historical pages that no longer behave as stable endpoints.
This is not unique to Italy, but Italy is a particularly good case study because the migration has real consequences for coverage.
Zero captured records does not mean zero filings existed. It means our current ingestion path is not reaching them.
eMarket STORAGE is an authorised storage mechanism for regulated information in Italy. In plain English, it is one of the places where public disclosures are stored and made available after dissemination. For users of insider dealing data, this means that the practical retrieval target is often a storage platform rather than the regulator’s own site.
That sounds harmless until one tries to build a historical panel.
A clean data regime has three traits:
Migrating publication away from a regulator-facing archive can weaken all three if the old records are not redirected, the new platform’s search interface is not scrape-friendly, or issuers vary in how they title and categorise filings.
Another problem is categorisation. Insider dealing announcements are often only one subtype within a broad universe of regulated information, alongside financial statements, voting-rights notices, governance updates, and bond documentation. If the storage system is not optimised for insider-dealing retrieval, users can face a familiar sequence of irritations:
For a human reader, this is survivable. For a data pipeline, it is a tax.
Current filings can usually be found with enough persistence. Historical continuity is harder.
The migration story creates at least three archival layers:
These layers do not always agree on permanence. A link may resolve today and fail next quarter. A PDF may remain available while its search page disappears. A filing may be discoverable by issuer but not by transaction type. For a cross-market database, this is the sort of mess that quietly becomes selection bias.
A source migration rarely announces itself in a way a scraper appreciates. The code usually expects one of three things:
When disclosures move from a regulator site to a storage platform, all three assumptions can fail at once. Even if the legal disclosure flow continues uninterrupted, the machine-readable continuity may not.
The immediate symptoms are familiar:
That first symptom is exactly what our Italian count shows.
Researchers often underestimate URL fragility. If a regulator-hosted page once linked to a disclosure, and that disclosure later lives behind a different search endpoint or storage interface, the old URL may not redirect to the new canonical record. Instead, it may produce one of the following:
This matters because historical datasets are often built from old references, not just current search results. Once the old path breaks, reproducibility goes with it.
Suppose you compare insider buying frequency across France, the UK, and Italy. France may offer relatively coherent public pathways through AMF-related disclosure channels and issuer publication under MAR. The UK has National Storage Mechanism records and broad dissemination through RNS-linked infrastructure. Italy may require a more fragmented retrieval process involving eMarket STORAGE and issuer archives.
If you do not adjust for source architecture, Italy can look artificially quiet. The market is not quieter. Your plumbing is.
| Market | Regulator | Rule | Deadline | Notes |
|---|---|---|---|---|
| FR | AMF | MAR Art 19 | T+3 | Public retrieval often begins with issuer disclosures and AMF-related market infrastructure; still requires issuer-level verification. |
| IT | CONSOB | MAR Art 19 | T+3 | Operational retrieval often runs through authorised storage mechanisms such as eMarket STORAGE rather than a single regulator-hosted insider-dealing archive. |
| US | SEC | Exchange Act Section 16 | T+2 | EDGAR provides a central, regulator-hosted archive with structured forms, making historical retrieval materially easier. |
The table is not there to flatter the SEC. Even EDGAR has its charms. But centralisation does reduce ambiguity.
A robust retrieval method begins from the disclosure obligation, not from whichever website seems official. For Italy, that means mapping the publication chain around MAR Article 19 and the Italian regulated-information infrastructure.
A practical search order is usually:
This is less elegant than a single regulator archive. It is also more likely to produce results.
Issuer IR pages are often the easiest place to find a current internal dealing announcement. They are also unreliable as a sole archive. Companies redesign sites, prune old releases, change CMS vendors, and occasionally treat permanence as a theoretical virtue.
Still, issuer pages are useful for:
The catch is consistency. One issuer may title the release “Internal Dealing”. Another may use “Operazioni di internal dealing”. Another may bury it under “Price Sensitive”. Machines dislike creativity.
For Italy, the sensible engineering response is to maintain a source map with explicit provenance fields:
That sounds bureaucratic. It is cheaper than pretending one URL is canonical and discovering later that half your historical links now point to nowhere in particular.
When legacy CONSOB references or old issuer links fail, web archives become part of normal research rather than a last resort. They can recover:
This does not restore a perfect dataset. It does, however, let you distinguish between “not found” and “not filed”. Those are very different statements, and only one of them belongs in a serious article.
Our current Italian count, zero declarations, is not merely a technical inconvenience. It is a governance signal for the dataset itself. Any platform claiming broad European insider coverage should be able to answer, market by market:
If those answers are vague, the coverage claim is vague too.
A market with zero captured records is not a market insight. It is a to-do list.
Many datasets describe themselves as pan-European because they cover MAR jurisdictions. That is legally true and operationally slippery. MAR harmonises obligations to a degree. It does not harmonise archive design, storage interfaces, or historical URL hygiene.
Italy exposes the difference between regulatory harmonisation and data harmonisation. The former is a legal project. The latter is mostly janitorial work with occasional SQL.
Academic and practitioner studies on insider trading often assume filing availability is sufficiently uniform to support pooled analysis. In reality, source architecture can influence observed frequency, timing, and even the apparent mix of buys versus sells if one category is easier to retrieve than another.
For Italy, any empirical work using public disclosures should state:
Without that, “Italy underperforms peers in filing intensity” may simply mean “Italy underperforms peers in archive convenience”.
CONSOB’s role as regulator is not in doubt. The practical issue is that users often conflate regulatory authority with archival centrality. Once publication flows through authorised storage mechanisms and issuer dissemination systems, the regulator’s domain may no longer be the complete public record in the way researchers expect.
That distinction should be made explicit in market guides, vendor documentation, and methodology notes. Otherwise users continue to search the wrong place, fail to find records, and infer a market pattern from an infrastructure quirk.
A serious rebuild of Italian insider-dealing coverage would probably include five steps:
That is not glamorous work. It is, however, the difference between a usable dataset and a decorative one.
Italy’s migration from a CONSOB-centred public perception to an eMarket STORAGE-centred retrieval reality is a reminder that disclosure systems are socio-technical objects. The law may say the filing exists. The archive decides whether anyone can still find it.
For readers tracking insider activity, the practical payoff is simple. If you are missing Italy, do not ask first whether insiders stopped trading. Ask where the documents went, which URLs died, and whether your source map still reflects the market as it actually publishes. The next concrete step is to build that map, issuer by issuer, and then test how far back eMarket STORAGE plus issuer archives can restore a continuous Italian series. The open question is not whether the filings exist in principle. It is how many remain discoverable in practice.
Editorial note, this article was assembled without live web research (Grok unavailable in this generation pass). Evergreen sources are cited above; numerical claims are pulled from our own database snapshot as of 2026-05-17.
The analysis of Italy's insider declarations data reveals notable insights into market behavior. Our backtest figures indicate that the BUY cohort, consisting of 130 priced trades, yielded a mean 90-day return of -2.48% with a win rate of 51.54%. This performance is contrasted by the broader global universe, which encompasses 227,594 BUY filings across 17 markets and shows an aggregate mean 90-day return of -1.70% and an aggregate Sharpe ratio of -0.08. These statistics underscore the challenges faced in leveraging insider declarations for predictive trading strategies in Italy, highlighting the need for further investigation into the underlying factors influencing these returns. It is important to note that the sample size and time horizon may limit the robustness of these findings.
Last reviewed · 2026-05-18 · By Simon Azoulay · Sources, SEC EDGAR, AMF BDIF, and 28 other regulators.
Brad Kitchen bought about EUR 24,012 of Element One stock as Canada’s natural hydrogen and critical minerals story keeps...
Don Gray bought Petrus Resources shares into a soft Canadian energy tape. Here is how the cluster reads against peers, o...
Predilife founder Stéphane Ragusa bought again as European life sciences stays cautious, with ALPRE still a micro-cap an...
Arqit’s CEO sold after a strong run in quantum names. The filing lands against weak revenue, a fresh cluster of sales, a...
Cardlytics CEO Amit Gupta sold near $4.39 after a reverse split. We read the cluster against a shaky ad-tech tape and sm...
GreenPower’s CEO bought about EUR 1.01m of stock as EV demand stays uneven. We read the filing against Workhorse and the...
