A deed of company arrangement (DOCA) is a binding agreement between a company in voluntary administration and its creditors, under which creditors accept a compromise of their claims in exchange for a better return than an immediate winding up. In Australia, deeds are governed by Part 5.3A of the Corporations Act 2001.
ASIC has now published four years of data on how that process actually performs. Report 836 covers every voluntary administration in Australia between 1 July 2021 and 30 June 2025: 5,020 companies, grouped into 3,528 appointments, carrying $71 billion in liabilities. It is the most detailed public account of the VA and DOCA process ASIC has produced.
Most of the commentary has settled on one number, around 44% of administrations end in a deed. The more useful material is in the failure analysis, and it has attracted almost no attention.
Key findings at a glance
- Of 1,100 finalised deeds, 893 (81%) were wholly effectuated and 191 (17%) failed and went into creditors' voluntary liquidation.
- Deeds that failed were far more likely to be funded from future trading profits: 63.3% of failures against 29.8% of successes.
- Whether the deed provided for the business to keep trading made almost no difference 68.7% of failures, 66.4% of successes.
- Deeds funded from trading profits took a median of 362 days to complete, against 195 days for those that were not.
- Of the proposals put to creditors, 86.6% were accepted. Even proposals estimating a nil return were accepted 66.4% of the time. Creditors wanted to keep trading with the company.
Source: ASIC Report 836, July 2026.
How often do deeds of company arrangement fail?
Of the 1,500 deeds approved during the review period, 1,100 had been finalised as at 31 May 2026 and 400 were still running.
Of the 1,100 finalised:
- 893 (81%) were wholly effectuated obligations under the deed fulfilled and all admissible claims dealt with
- 191 (17%) entered creditors' voluntary liquidation
- 16 (2%) were finalised with another outcome
On completed deeds, only one in six failed after the deed was approved.
That figure should be read with care. Deeds tend to fail late rather than early, and 400 deeds in this population had not yet finished when ASIC drew its data. The eventual failure rate across the full cohort is not yet known, and the 17% is more likely to understate it rather than overstate it.
What separated the deeds that failed?
ASIC compared the 191 deeds that passed into liquidation against the 893 that were wholly effectuated. This is the single most useful table in the report.
| Deeds that failed | Wholly effectuated | |
|---|---|---|
| Number of deeds | 191 | 893 |
| Funded from trading profits | 63.3% | 29.8% |
| Included a third-party contribution | 50.3% | 68.4% |
| Median liabilities | $3.5m | $2.7m |
| Median deed fund | $511,990 | $364,630 |
| Deed provided for continued trading | 68.7% | 66.4% |
Source: ASIC Report 836, Table 12·Percentages are shares of each outcome group.
The funding source is the dividing line
Read the last row of that table first. Whether the deed provided for the business to keep trading was almost identical across both groups, 68.7% of the failures against 66.4% of the successes. On this evidence, continued trading is close to irrelevant to whether a deed survives.
What differed was where the money was coming from. Deeds funded out of future trading profits made up 63.3% of the failures and 29.8% of the successes. Deeds with a third-party contribution made up 50.3% of the failures and 68.4% of the successes.
ASIC reports these as shares of each outcome group rather than as failure rates. Turning them around gives a sharper picture. Across the 1,084 finalised deeds in those two groups, our calculation from ASIC's figures is:
- Deeds funded from future trading profits: approximately 31% failed
- Deeds funded from other sources: approximately 10% failed
Around three times the failure rate. This is our derivation from ASIC's published percentages, not a figure ASIC reports. It excludes the 16 finalised deeds with other outcomes and the 400 deeds still running.
Why funding from trading profits is the weak point
ASIC's own explanation is short and worth quoting in full. "DOCAs that depended on future trading profits generally took longer to complete and were more likely to fail and enter liquidation. This is not surprising, given that creditor returns are less certain in these circumstances. Where a DOCA depends on future trading profits, the viability of the business and the assumptions supporting future profitability are particularly important."
The mechanics are straightforward. A deed funded by a third-party contribution is, in substance, funded at the outset 935 deeds (63%) included a contribution, worth just over $1 billion in new money, and where a contribution was made it represented 86% of the deed fund on average. In 49% of those appointments, it was the entire fund. The money either arrives or the deed does not proceed.
A deed funded from future trading profits is a forecast. It requires the business to perform, over a period, in conditions that have already proved difficult enough to put it into administration. Every month of underperformance is a month the deed is not being funded.
The duration data shows the exposure. Deeds funded from trading profits ran to a median of 362 days; those funded otherwise, 195 days. Failed deeds also ran longer than successful ones overall - an average of 369 days and a median of 290, against 316 and 238 for those that wholly effectuated. A deed that depends on trading is exposed for roughly twice as long to the risk that trading disappoints.
Continued trading is not the risk. Funding the deed out of it is.
This distinction matters commercially, because the two are separable and are frequently conflated.
Of the 730 deeds where the business continued to trade after execution, only 320 (44%) included a contribution from future trading profits. The other 410 (56%) had the business trade on while the deed itself was funded from somewhere else a shareholder or related-entity contribution, an external investor, or an asset realisation.
That combination the business continues, but the creditors' return does not depend on it is both available and common. On ASIC's data it is the stronger structure, and it is the one we would build towards wherever a contribution can be found.
A bigger deed fund is not a safer one
The deeds that failed were the larger ones. Median liabilities of $3.5 million against $2.7 million, and a median deed fund of $511,990 against $364,630. The instinct that a larger contribution makes a proposal safer is not supported here.
Two cautions on reading, ASIC does not claim the size of the fund causes failure, and size travels with complexity related-entity groups, secured creditors, cross-guarantees, creditors' trusts. Larger appointments are harder for reasons that have nothing to do with the fund.
Funding from trading profits was more common at the smaller end around 50% of continuing-trade deeds where liabilities were between $500,000 and $5 million, against 22% where liabilities exceeded $10 million. So, the weaker funding structure is concentrated among smaller companies, while the failures skew larger. Both are in the data.
What this means when structuring a proposal
Four working conclusions we draw from the failure analysis:
- Find the contribution before drafting. Where third-party money is available, it changes the risk profile of the deed more than any other variable.
- Do not conflate trading on with funding from trading. More than half of continuing-trade deeds were funded independently of trading performance, and that is the structure to aim at.
- Where trading profits must fund the deed, the forecast is the deed. The viability of the business and the assumptions supporting future profitability are "particularly important" and is a direct invitation to stress-test them before creditors do.
- Shorter is safer. A deed structured to complete in months rather than years is exposed for less time to the thing most likely to break it.
Is voluntary administration being used less?
As a share of external administrations, yes. Voluntary administrations accounted for 35–40% of all external administrations in Australia from FY00 to FY06, around 15% through the 2010s, and around 10% in FY25 and FY26.
In absolute terms appointments have risen 502 in FY22 to a peak of 1,077 in FY24, easing to 1,071 in FY25 while the share has kept falling, alongside the growth of the small business restructuring process.
The size of company using it has not shifted much. The proportion of appointments with less than $1 million in liabilities moved only between 24.6% and 29.8% across the four years. ASIC is careful here: the data does not show a marked move away from smaller VA appointments, and the degree of substitution between SBR and VA cannot be determined from it.
The median appointment carried $2.34 million in liabilities. Construction accounted for a quarter of all appointments (886) and a third of all liabilities ($23.6 billion).
Why do half of all administrations end without a proposal?
At least 96% of appointments reached a second creditors' meeting. Of those, 1,684 had a DOCA proposal in front of creditors and 1,681 did not. Where no proposal was put, 93% of those companies went into creditors' voluntary liquidation.
The pattern is concentrated at the smaller end. Fewer than one in three appointments with liabilities under $1 million produced an approved deed, and ASIC notes that the majority of those ended in liquidation with no proposal put to creditors at all.
On that cohort the report says the position "raises questions about the circumstances in which VA is being used where no proposal is ultimately put to creditors" suggesting this may in some cases reflect directors seeking advice or taking action only after the financial position had materially deteriorated, and in others that a different form of external administration may have been more appropriate.
Does the timing of an appointment matter?
There were winding-up proceedings within 90 days before appointment in 388 appointments (11%). Of those, 227 (59%) went into liquidation. Around 26% proceeded to a deed, against 45% where no winding-up application had been made.
ASIC reports the association without explaining it, and the report does not establish that the filing itself changes the outcome a company facing a wind-up application is likely to be further along in its distress either way. But the difference is large, and it is the clearest signal in the report that when there is a winding up application, deeds of company arrangement are more likely to fail.
Does company size determine whether a deed is approved?
Larger appointments were more likely to produce a deed: 48.3% above $10 million in liabilities, against 15.4% between $1 and $250,000. But the transition rate does not climb steadily with size. It rises sharply through the sub-$1 million bands and then flattens 49.7% for $1–2 million, 47.9% for $2–5 million, 54.5% for $5–10 million and 48.3% above $10 million.
The threshold that matters is much lower. Between $500,000 and $1 million the rate is still 36.6%, and between $250,000 and $500,000, 33.8%. It is below $250,000 that it collapses, to 15.4%.
Will creditors accept a modest proposal?
Of the proposals actually put to creditors, 86.6% were accepted. Acceptance is high across the whole range of offers. Proposals carrying a high estimate of 1–10 cents in the dollar were accepted 87.7% of the time; those estimating 100 cents, 92.7%. Even proposals estimating a nil return to unsecured creditors were accepted in 66.4% of cases. Only 16% of proposals estimated a return above 50 cents.
The rate does climb modestly with the size of the offer, so a bigger number does not hurt. The point is the floor rather than the slope: a proposal offering ten cents is not, on this evidence, facing a materially different reception from one offering a dollar.
Administrators recommended creditors accept 95% of the proposals they received, and creditors followed that recommendation 89% of the time. Of the 5% administrators did not recommend, 42% were approved anyway.
The constraint is not creditor appetite. It is getting a fundable proposal in front of them at all.
What do creditors actually receive under a deed?
At proposal stage, administrators recorded a high estimate averaging 26 cents in the dollar (median 15 cents), where "high" means the top of an estimated range rather than a generous return. Across wholly effectuated deeds, the average actually paid to unsecured creditors was 21.3 cents, with a median of 11.5 cents.
These are different populations, every proposal on one side and completed deeds excluding creditors' trust matters on the other, so the pair is not a like-for-like measure of shortfall. Data on dividend distributions under the Creditors Trust are not reported to ASIC.
Where a VA ended in creditors' voluntary liquidation, administrators estimated that only 26% of those liquidations might pay any dividend at all. Averaged across all of them the high estimate was 9 cents in the dollar, and the median was nil.
Almost 90% of the 712 wholly effectuated deeds that reported a final return outside a creditors' trust paid a dividend to unsecured creditors, totalling $238 million. Across all deeds in the review period, $693.8 million was distributed: $315.9 million to unsecured creditors, $269.5 million to secured creditors, $80.3 million to employees and $28.0 million to contributories.
Employee priority claims fare comparatively well. Among the 413 wholly effectuated deeds outside a creditors' trust that had priority wage and superannuation creditors, 80.1% paid a dividend at a median rate of 100 cents in the dollar. The picture is less uniform elsewhere: priority leave claims were paid in 39.6% of the deeds that had them and retrenchment claims in 65.9%, both also at a median of 100 cents where paid.
What does a creditors' trust remove from the record?
Every return figure above carries a qualification that is easy to miss. A large part of the deed population is not in them.
A creditors' trust is a structure used alongside a deed. Instead of paying creditors under the deed itself, the company transfers the deed fund and the creditors' claims into a separate trust. The deed is then effectuated, and the company is released from external administration, often within weeks. Creditors become beneficiaries of that trust and are paid, if they are paid, by the trustee outside the deed. ASIC's expectations for the use of creditors' trusts are set out in Regulatory Guide 82.
The structure is lawful and sometimes necessary. What it also does is end the reporting. Once the deed is effectuated the trust sits outside the Part 5.3A framework, and distributions made inside it are not lodged with ASIC.
How much of the deed population sits inside a trust?
A creditors' trust was used in 172 of the deeds ASIC reviewed, around 12% of approved deeds. Counted that way it looks marginal. Counted by value it is not.
Those 172 appointments covered 376 companies. On our arithmetic from ASIC's published figures, they carried around $14.8 billion in liabilities, roughly 31% of all liabilities in the deed population, including around $8.6 billion of unsecured creditor claims, roughly 47% of unsecured creditor value across all deeds. Their median liabilities were $11.86 million, against $2.38 million for deeds without a trust.
ASIC reports that 36% of the 113 deeds with liabilities above $10 million that provided for continued trading used a creditors' trust, and that use varies markedly by sector, with mining the highest at around 53%.
The 31% and 47% figures are our calculations from the REP 836 data pack, not figures ASIC publishes.
What the reporting shows
| No creditors' trust | Creditors' trust used | |
|---|---|---|
| Number of deeds | 1,328 | 172 |
| Reported any dividend to ASIC | 53.6% | 14.0% |
| Median total dividend reported | $109,329 | Nil |
| Median deed remuneration, finalised | $11,000 | Nil |
| Median liabilities | $2.38m | $11.86m |
| Median deed duration | 293 days | 19 days |
| Recorded as wholly effectuated | 79.8% | 94.4% |
| Included a third-party contribution | 60.8% | 77.3% |
| Funded from trading profits | 46.1% | 23.9% |
| Return to unsecured creditors, where reported | 21.7% | 21.8% |
Olvera analysis of the ASIC REP 836 data pack. Dividend and remuneration medians are of amounts reported to ASIC·Duration for trust matters is 18.5 days on the data pack and is reported as 19 days in the report.
Why a 94.4% effectuation rate is not a 94.4% success rate
A creditors' trust was used in 172 of the deeds ASIC reviewed, around 12% of approved deeds. Counted that way it looks marginal. Counted by value it is not.
Those 172 appointments covered 376 companies. On our arithmetic from ASIC's published figures, they carried around $14.8 billion in liabilities, roughly 31% of all liabilities in the deed population, including around $8.6 billion of unsecured creditor claims, roughly 47% of unsecured creditor value across all deeds. Their median liabilities were $11.86 million, against $2.38 million for deeds without a trust.
ASIC reports that 36% of the 113 deeds with liabilities above $10 million that provided for continued trading used a creditors' trust, and that use varies markedly by sector, with mining the highest at around 53%.
The 31% and 47% figures are our calculations from the REP 836 data pack, not figures ASIC publishes.
Our reading: a deed can be wholly effectuated without any money reaching creditors under the deed, because the obligation performed is the transfer of claims into the trust. The 94.4% effectuation rate for trust matters should be read as a measure of structure, not of outcome.
What the data does not show
It does not show that creditors in trust matters do worse. Where trust deeds did report a return, the aggregate rate to unsecured creditors was 21.8%, effectively identical to the 21.7% reported by deeds without a trust.
The difficulty is how little evidence there is. For 86% of trust matters no dividend is reported to ASIC at all, so the comparison rests on the minority that do report. Whether the other 86% paid nothing, paid well, or paid late is not knowable from this dataset.
What it does to the headline numbers
Four adjustments a reader should carry into the rest of the report:
- ASIC excluded 145 trust matters from its analysis of dividends. The return figures in this article, including the 21.3 cent average and 11.5 cent median to unsecured creditors, describe deeds outside a trust.
- The same applies to the 712 deeds that reported a final return and the 413 deeds with employee priority claims. Both populations are non-trust.
- Effectuation and failure rates are not adjusted. The 81% effectuated and 17% failed split includes trust matters, where effectuation means something different.
- Duration figures are pulled down by the trust cohort. A 19-day median sits inside the overall averages.
None of this is hidden. ASIC states the exclusions. But the qualifications are scattered through the report, and the effect of putting them together is that the deed population as commonly described is smaller and more homogeneous than the headline count suggests.
Why it matters when advising on a proposal
A creditors' trust does real work. It releases the company from external administration quickly, which can preserve licences, contracts, banking and the goodwill that a prolonged deed erodes. For a large trading group, that speed is often the reason the restructure is viable at all.
Our position is that the structure should be judged on what it discloses, because the public record goes quiet from the moment the deed effectuates. Where a proposal uses a trust, the questions worth putting are who the trustee is and how they are remunerated, what the trust will actually hold and when, what the expected distribution timetable is, and what reporting beneficiaries will receive once ASIC's reporting stops. A trust proposal that answers those four questions in the administrator's report to creditors is a different proposition from one that does not.
What does the process cost?
The median cost of a voluntary administration was around $68,000 (average $206,000). The median cost of a deed was around $33,000 (average $107,000). For the 838 wholly effectuated deeds that reported both, the median total across the whole process was around $111,000.
Complexity is the main driver. A single-company voluntary administration ran to a median of $60,000 in approved remuneration; one covering eleven or more related companies, around $749,000. The equivalent deed figures were $28,970 and $247,500.
Outcome matters too, in the direction you would expect: administrations that produced a deed carried median remuneration of around $88,000, against $62,000 for those that ended in creditors' voluntary liquidation. A deed is the more expensive path, which is worth stating openly, alongside the fact that the alternative was estimated to return a median of nil.
What we take from it
Structure decides survival more than circumstance does. The clearest signal in four years of data is that deeds relying on future trading profits made up 63.3% of failures and 29.8% of successes, while whether the business kept trading barely moved the needle. Where a proposal can be funded independently of trading performance, it should be.
Timing is the one variable a director still controls. Appointments made before a creditor has moved sit in a materially better part of this dataset, 45% against 26%. The report cannot tell us how much of that gap moving earlier would close, since a company facing a wind-up application is further into its distress either way. But it is the only input on this list wholly within a director's gift.
Do not assume a modest offer is a weak one. Creditors accepted proposals carrying a high estimate of one to ten cents 87.7% of the time. What the report does not tell us is whether modest proposals complete more often, but pitching an offer the business can actually fund is the one part of a proposal a director controls entirely.
Get a proposal in front of the meeting. Almost half of all administrations arrive at the second creditors' meeting with nothing to vote on, and 93% of those end in liquidation.
Read a trust proposal on its disclosure, not its effectuation rate. Deeds using a creditors' trust record a 94.4% effectuation rate and a median reported dividend of nil, because effectuation only requires the claims to move into the trust. Around 12% of deeds used one, but on our arithmetic, they carry roughly 31% of the liabilities in the deed population. Where a proposal uses a trust, the trustee, the timetable and the reporting to beneficiaries are what a creditor should be weighing.
Frequently Asked Questions
What is a deed of company arrangement?
A deed of company arrangement is a binding agreement between a company in voluntary administration and its creditors, under which creditors accept a compromise of their claims in return for a better outcome than an immediate winding up. It is governed by Part 5.3A of the Corporations Act 2001 (Cth) and takes effect once creditors resolve to accept it at the second creditors' meeting.
How often do DOCAs fail?
Of the 1,100 deeds finalised in ASIC's 2021–2025 review population, 191 (17%) passed into creditors' voluntary liquidation and 893 (81%) were wholly effectuated. Because 400 deeds in the population were still running when ASIC drew its data, and deeds tend to fail late, the eventual failure rate may be higher.
Why do DOCAs fail?
The strongest differentiator in ASIC's data is the funding source. Deeds funded from future trading profits made up 63.3% of failures but only 29.8% of successes, while deeds including a third-party contribution made up 50.3% of failures and 68.4% of successes. Whether the business continued trading made almost no difference.
What percentage of voluntary administrations end in a DOCA?
Around 44% of voluntary administrations in ASIC's review population resulted in creditors accepting a deed. Of appointments that reached a second creditors' meeting, half had a proposal in front of creditors, and 86.6% of those proposals were accepted.
How much do creditors receive under a DOCA?
Across wholly effectuated deeds, unsecured creditors received an average of 21.3 cents in the dollar, with a median of 11.5 cents. Almost 90% of those deeds paid a dividend. Where a voluntary administration instead ended in liquidation, administrators estimated a median return to unsecured creditors of nil.
How long does a DOCA take?
Finalised deeds ran to an average of 327 days and a median of 248 days. Deeds funded from future trading profits took considerably longer, a median of 362 days against 195 days for deeds funded from other sources.
What is a creditors' trust, and why does it matter?
A creditors' trust is a structure used with a deed of company arrangement under which the deed fund and creditors' claims are transferred into a separate trust, allowing the deed to be effectuated and the company released from external administration quickly. Creditors are then paid by the trustee outside the deed. In ASIC's data, 172 deeds used one. Only 14% reported any dividend to ASIC, because distributions made inside a trust fall outside ASIC's reporting framework.
Does a winding-up application affect the chance of a DOCA?
In ASIC's data, appointments preceded by winding-up proceedings within 90 days produced a deed around 26% of the time, against 45% where no application had been made. ASIC reports the association without establishing that the application itself causes the difference.
Source: ASIC Report 836, Review of voluntary administration and deed of company arrangement process: 2021–2025, published 7 July 2026, together with the accompanying REP 836 data pack. ASIC notes that its figures represent an analysis of available lodged data rather than a complete assessment of every appointment, that some lodged forms contained errors or misclassifications, and that the data will not reconcile to ASIC's published insolvency statistics.
This article is general information only and is not legal, financial or insolvency advice. Figures describe ASIC's review population and do not predict the outcome of any particular administration. You should obtain advice specific to your circumstances before acting.
