Composition and creditor voting: cost and likely outcome hinge on three factors: how creditors split into classes, what share of each class backs the proposal, and how the administrator's fee tracks the size of the estate. There is no fixed price and no guaranteed result. The vote either clears the required majority in every class, or it does not.
Who this concerns
This concerns a company that has entered formal reorganisation and put a composition to its creditors, and every creditor named on the claims list who has to decide how to vote. It also concerns a creditor who expects to vote but is not listed, because being left off the register changes the outcome as much as a rejected vote does.
The situation is not symmetrical. The board authorised the proposal and carries personal exposure that a successful vote does not automatically clear. A secured creditor typically sits outside the vote, unless part of its claim is unsecured or disputed, in which case that portion is voted like any other unsecured claim. All of this sits inside the wider process covered by the insolvency and restructuring practice, where the composition is one stage among several rather than a standalone event, and the vote is rarely the first decision that matters, only the most visible one.
What the law says
Swedish reorganisation law lets a debtor propose a composition once formal reorganisation has opened, and settles the outcome by vote rather than by case-by-case negotiation with each creditor. Under Swedish law as it currently stands, claims are grouped into voting classes, a majority is required inside each class, and a class that rejects the proposal blocks it regardless of how strongly other classes support it.
The administrator convenes the vote, counts it by number of creditors and by value of claims within each class, and reports the result to the court. Secured claims sit apart from unsecured claims, and a disputed claim votes conditionally until the dispute is resolved one way or the other. None of this cancels the debt if the vote fails: it means the reorganisation continues without the reduction the composition would have delivered, and liquidation becomes the likelier next step, not an automatic one.
How it works in practice
How the proposal reaches a vote
The administrator drafts the proposal once the claims list is reasonably settled, sets out what each class is offered, and files it with the court together with a summary of the estate. The court sets a date for the vote, and the proposal, together with the claims list, is made available to every creditor with enough time to review it before that date. A creditor who wants to contest its own classification, or the size of another creditor's claim, needs to raise it before the vote, not after, because the tally is taken at the meeting itself and is not reopened afterward as a courtesy.
Who gets a vote and who does not
Unsecured creditors vote in full. Secured creditors generally do not vote on the secured portion of their claim, because the composition affects unsecured recovery, not the security itself. A creditor whose claim is contested by the administrator, or by another creditor, is admitted to vote provisionally, and the final effect of that vote depends on how the dispute is resolved afterward. A creditor omitted from the list entirely has to apply to be added before the vote, and a late application is not guaranteed to succeed in time to count.
How classes are formed
Classes are formed by the nature of the claim, not by the identity of the creditor. Trade creditors, tax claims, and employee claims are typically kept apart because they carry different priority and different appetite for a write-down. A class with one dominant creditor behaves differently from a class made up of many small claims of similar size, because a single creditor can decide the class outcome alone, while a fragmented class needs coordinated support from several unrelated parties to reach the same result.
What changes if a creditor is based outside Sweden
A foreign creditor votes on the same terms as a domestic one once its claim is admitted, but getting there is slower. Notice has to reach it in time to matter, translation of the proposal and the claims list is often necessary in practice even where not formally required, and a foreign creditor's own internal approval process for accepting a write-down can outlast the window the Swedish timetable allows. Where the foreign creditor's security is governed by a law other than Swedish law, that security is not restructured by the Swedish vote at all, and the domestic process proceeds around it rather than through it, which changes what the vote actually settles for everyone else.
What drives the administrator's fee
The fee tracks the size and complexity of the estate, not the size of the company's turnover. A claims list with many disputed items, a creditor base spread across several classes, or a business with assets that need separate valuation all add hours before the vote can even be called. A composition proposed early, on a settled claims list, with a small number of classes, costs less to bring to a vote than one proposed after months of dispute over who owes what to whom.
What drives legal cost for the company
For the debtor company, cost concentrates before the vote: preparing the claims list, responding to disputes raised by creditors, and briefing the administrator on assets that are hard to value. Once the vote is set, cost is largely fixed, because the proposal itself cannot be renegotiated creditor by creditor in the days before the meeting, and reopening it at that stage tends to cost more than it saves.
What drives legal cost for a creditor
For a creditor, cost tracks the size of the claim relative to its class. A claim large enough to decide the class outcome on its own justifies close review of the estate and the proposal; a small claim inside a large class rarely does, because one vote among many changes little regardless of how carefully it is reasoned or how thoroughly the estate is checked beforehand.
What to check before the vote
- Which class the claim sits in, and whether that classification matches the nature of the claim rather than an administrative default
- Whether the claim is listed at its full value, and whether any part of it is treated as secured when it should not be
- What proportion of the class the claim represents, and whether that proportion is enough to swing the class result on its own
- Whether any other creditor in the same class has a connection to the debtor that could affect how it votes
- Whether the proposal treats claims of the same priority consistently across the different classes
- Whether a foreign security interest is properly excluded from the vote rather than assumed to be covered by it
- What happens to personal liability exposure for board members if the vote succeeds, since a successful composition does not resolve that question separately
Which court or authority decides a liability claim against a de facto director?
A claim that someone acted as a de facto director, rather than a board member of record, is decided by the general courts in the same way as a claim against a formally appointed director, because Swedish law looks at the function performed rather than the title held. Running a composition process does not change which court hears that claim, or move it away from the ordinary civil track, and a pending composition is not a defence against it.
If employees become suspects in a related investigation, what is the company's position and what deadlines apply?
The company's position depends on whether it is treated as a suspect itself, as a victim, or as a third party holding evidence, and that classification changes what it has to disclose and by when. Deadlines in a criminal investigation run independently of the reorganisation timetable, and a company cannot use a pending composition to delay a response it would otherwise owe to an investigating authority.
How does asset tracing and recovery work when a creditor needs to reach assets in Romania?
Recovery against assets located in another EU member state generally follows the recognition route available under the instruments that apply between Sweden and that state, rather than a fresh domestic claim brought from scratch. Whether that route is faster than pursuing a Swedish claim to judgment first depends on where the debtor's assets actually sit and how contested the underlying claim is, which is why tracing usually precedes any decision on how to enforce.
The numbers
The majority required in each class, the notice period before the vote, and the filing deadlines are all fixed by the applicable provisions rather than agreed case by case, so they are confirmed at the point of filing rather than assumed here. What can be said without a specific figure is the direction of movement: cost rises with the number of voting classes, the number of disputed claims still open when the vote is called, and the length of time any single class spends in opposition before the dispute clears.
Timing follows the same pattern. It is set by the court's calendar and by how quickly disputed claims are resolved, not by a standard number of weeks that applies across cases regardless of complexity. For the company, the practical driver is the administrator's engagement with the size and complexity of the estate, not a fee scaled to turnover. For a creditor, the driver is usually the size of the claim relative to its class, because a claim large enough to decide the class outcome justifies more scrutiny of the estate than a small one does, and that scrutiny itself has a cost that scales with what is at stake.
Where it usually goes wrong
If secured creditors hold enough of the debt to make the unsecured class immaterial to the overall outcome, the vote becomes close to symbolic, and the real negotiation happens with the secured side outside the process entirely, on terms the vote never touches.
If a single creditor holds a blocking share within one class, the proposal fails at that class regardless of overwhelming support elsewhere, and there is no mechanism to override one class's rejection with the support of another.
If assets or a parent company sit outside Sweden, and a foreign creditor's security is governed by a law other than Swedish law, the domestic vote cannot restructure that security. The process proceeds around it, and what other creditors eventually recover is set by whatever remains once that claim is satisfied first, not by the percentage stated in the proposal.
Where the claims list itself is disputed, voting stalls until the dispute is resolved, and a creditor that assumed its vote would count can find the meeting closed before the dispute is settled in its favour.
A composition that clears the vote does not by itself release board members from personal liability claims that arose before the reorganisation began, and it does not pause a parallel investigation into how the company reached this position in the first place.
If the vote fails, the file does not return to where it started. It moves toward liquidation with less runway and a shorter list of options than existed before the proposal was ever put forward.
What to do next
Model the vote before it is called, not after: map the classes, the size of each relative to the whole, and which creditors are large enough to decide a class on their own. That mapping tells a company whether the proposal on the table is realistic, and tells a creditor whether its vote actually matters or whether the outcome is already decided by someone else's claim.
This is where the analysis in the trustee's investigation into the board becomes relevant, if the composition vote is only one part of a wider review of how the company reached this position. Where the open question is narrower, whether this specific proposal is likely to clear and at what cost to reach that vote, an assessment call is the faster route: get in touch and bring the claims list and the draft proposal, and the call will focus on where the classes actually sit rather than on the process in general.