Thirteen Years of MR01: What Companies House's Reformed Charges Register Reveals About UK Secured Lending

On 6 April 2013, the Companies Act 2006 (Amendment of Part 25) Regulations swept away ninety-three years of accreted charge-registration practice and replaced it with a streamlined, UK-wide regime built around a single prescribed form: the MR01. Form 395, the dog-eared mainstay of corporate lending paperwork since 1948, was retired overnight. The change was made with minimal political theatre and almost no press coverage. Thirteen years on, it remains one of the quietest successes of UK corporate law reform — and one of the few corners of the Companies House estate that the Economic Crime and Corporate Transparency Act 2023 (ECCTA) has barely touched.

For a register that processes more filings each year than the entire PSC framework, the charges system attracts strikingly little editorial attention. That is partly because the parties who care most — secured lenders, insolvency practitioners and corporate counsel — already understand it intimately. But the data sitting on the open register reveals a great deal about how UK businesses borrow, what they pledge, and how cleanly the post-2013 system has worn in.

What the 2013 reform actually changed

The pre-2013 regime was a patchwork. Sections 860 to 877 of the Companies Act 2006 (themselves descended from the Companies Act 1948 via the 1985 consolidation) required charges to be registered only if they fell within a prescribed list of categories — charges on land, on book debts, on goodwill, on uncalled share capital and so on. Charges outside the list were unregistrable, and the registrar's certificate of registration was conclusive even where the underlying filing was inaccurate. Scotland operated a near-identical but legally distinct register, with its own forms and its own quirks around the moment of creation.

The 2013 reforms collapsed all of this into a single, optional, all-property regime under sections 859A to 859Q. The headline changes were:

• A single UK-wide system. The English/Welsh and Scottish registers were harmonised. The same form, the MR01, now applies to charges over property situated anywhere in the United Kingdom.
• No more categories. Almost any charge created by a company is registrable. The old debates over whether a particular security fell inside or outside the statutory list ended.
• A 21-day clock from the date of creation. This was kept, but the consequences of missing it were clarified: the charge becomes void against a liquidator, administrator and creditors, but remains valid between the parties (s859H).
• A certified copy of the instrument must be filed. This replaced the old practice of submitting particulars only, and put the actual security document on the public file — redacted, in practice, only for personal information.
• Registrar discretion ended. The registrar must register any MR01 that meets the formal requirements. The conclusive-certificate problem disappeared with it.

The drafters of the reform — principally the Department for Business, Innovation and Skills working with the City of London Law Society — got the balance broadly right. The system has needed almost no patching since.

Volumes on the register

The MR01 may not be the highest-volume filing at Companies House — that title belongs to the confirmation statement, with roughly four million CS01s a year — but it is, on a transaction-weighted basis, easily the most economically consequential. The table below uses Companies House annual-report figures, rounded to the nearest thousand, to give a sense of scale.

Two features stand out. First, the volume of new registrations is remarkably stable: it sat between 320,000 and 400,000 a year throughout the 2010s and has not strayed from that range since. Second, satisfactions (MR04 filings indicating that a charge has been discharged) have crept up as a proportion of new registrations — closing the gap between gross and net additions and suggesting that the register is, slowly, becoming more current rather than more silted.

Who is actually filing

The register's heavy users are familiar names. Sampling the open data product for a single month in early 2026 — the JSON feed published under the Companies House MR01 dataset — clearing banks (Lloyds, NatWest, Barclays, HSBC and Santander) account for roughly half of all registrations by volume. The asset-finance arms of those same groups, plus specialist invoice-discounters such as Bibby Financial Services and Aldermore, push the total to around 70 per cent. The remainder is a long tail of intercompany debentures, private-credit funds, bridging lenders and, increasingly, fintech-lender SPVs.

This concentration matters because the register's quality is, in practice, set by the discipline of its largest filers. Clearing banks have automated MR01 submission through their panel solicitors and rarely miss the 21-day window. The long tail — and especially intercompany debentures created in haste at the year-end — generates a disproportionate share of late-filing court applications under section 859F (relief from failure to register).

Fixed, floating and the Enterprise Act drift

The MR01 does not require the filer to classify the charge as fixed or floating, but the certified instrument almost always does. Twenty-three years after the Enterprise Act 2002 effectively abolished administrative receivership for floating charges created on or after 15 September 2003, the floating charge has not disappeared — but its character has changed.

The Enterprise Act introduced two design features that have shaped every debenture drafted since:

1. The prescribed part for unsecured creditors (s176A Insolvency Act 1986), which carves out a percentage of floating-charge realisations for unsecured creditors. The cap rose from £600,000 to £800,000 in April 2020 and has not moved since.
2. The qualifying floating charge holder's right to appoint an administrator out of court (Schedule B1 to the Insolvency Act 1986), which replaced administrative receivership as the lender's primary control tool.

The practical drift is that modern debentures use fixed charges over as many specific asset categories as possible — book debts, intellectual property, shares in subsidiaries — and reserve the floating charge as a sweeper. The 2005 House of Lords decision in Re Spectrum Plus sharply limited the ability to take a fixed charge over a fluctuating ledger of book debts, but careful drafting around control over the proceeds account has restored much of the position in practice. The result is that the register today contains a great many "composite" debentures whose effective character only emerges on insolvency.

The 21-day clock and the relief application

The 21-day window for registering a charge is short by international standards (Delaware allows no such fixed period; Australia allows 20 business days) and unforgiving. Missing it does not invalidate the charge between the parties, but it strips the lender of priority against a subsequent liquidator or administrator. The remedy is a section 859F application to the court for an order extending time — typically granted where the omission was accidental, no third-party rights have intervened, and the application is made promptly.

Reported figures from the Insolvency Service and the Companies Court suggest somewhere between 250 and 400 such applications a year. The number has been remarkably stable, and the success rate is high; the cost (court fees, counsel, the lender's solicitor) is not. The risk-adjusted lesson for lenders is the same it has always been: file early, file completely, and do not rely on the indulgence of the court.

The satisfaction problem

The register's most persistent quality issue is not at the front door but at the exit. Filing an MR04 to record that a charge has been satisfied is optional. Many lenders simply do not bother once the underlying facility has been repaid, particularly where the borrower is solvent and the security has lapsed in practice. The result is that any due-diligence search at Companies House returns a long list of historic charges that are, commercially speaking, dead — but legally still on the file.

The data hints at the scale of the problem. Cumulative MR01 filings since April 2013 run into the millions. Cumulative MR04 satisfactions are roughly two-thirds of that figure. The remaining third is a mix of charges that are genuinely still live and charges that have been quietly forgotten. Solicitors conducting acquisition due diligence routinely ask sellers to procure deeds of release for charges that the seller had no idea were still showing.

What ECCTA did not touch

The most striking thing about ECCTA, from a charges perspective, is how little of it engages with Part 25 at all. The headline reforms — director identity verification, the registrar's new objection powers, the registered-email regime, the lawful-purpose statement on incorporation — all sit elsewhere. The Act gives the registrar broader powers to query and remove material, but those powers have not so far been exercised aggressively against the charges register. The cost-recovery fee uplift in May 2024 raised the MR01 fee from £15 to £24 (paper) and £15 to £15 (electronic — unchanged), but the underlying regime is untouched.

That is, on the whole, the right answer. The 2013 reform settled the architecture, and the data suggests it has worn in cleanly. If a reform 2.0 is ever attempted, the obvious targets are not statutory but operational: a mandatory MR04 within, say, six months of facility repayment; a satisfaction-by-default mechanism for charges over a defined age with no recent activity; and proper machine-readable structuring of the certified-instrument data so that fixed-versus-floating, asset-class and lender-identifier fields can be extracted without OCR.

None of that is glamorous. Neither is the MR01. But thirteen years of stable filings, falling drift between gross and net additions, and an almost-complete absence of litigation about the register itself together describe a piece of legal infrastructure doing exactly what it was designed to do — and quietly enough that almost no one has noticed.