Boverket’s Credit Guarantee, Capital Adequacy and the Financing of Residential Construction

Background

Residential construction in Sweden remains materially below what is needed. Against that backdrop, Boverket – the Swedish National Board of Housing, Building and Planning – offers an important policy instrument in the form of its credit guarantee, intended to facilitate the financing of new housing by allowing the State to assume part of the credit risk. Where financing conditions are tight, the allocation of risk among borrowers, lenders and the State becomes critical. For banks, however, a reduction in economic credit risk is not, by itself, sufficient. The key question is whether the guarantee is structured in a manner that also reduces capital requirements under the applicable prudential framework.

Boverket’s credit guarantee may be granted to lenders that, in practice, will typically be subject to supervision by the Swedish Financial Supervisory Authority, such as banks, credit market companies and other credit institutions. The guarantee also presupposes framework agreements with Boverket. It is therefore not formally limited to banks as a category. In practice, however, it presupposes an identifiable counterparty that can enter into a framework agreement with Boverket and carry the credit risk on its own balance sheet.

When Does a Guarantee Provide Regulatory Capital Relief?

For a guarantee to be recognised under the CRR, that is, the EU Capital Requirements Regulation, it is not sufficient that it improves the credit profile in general terms. In the case of guarantees, Articles 213 and 215 are particularly relevant. In short, the guarantee must be (i) legally effective, (ii) clearly defined and (iii) practically usable.

If all of these conditions are satisfied, the guaranteed portion of the exposure may, in principle, be treated under the so-called substitution approach. In broad terms, this means that, for the purposes of calculating capital requirements, the bank may rely on the guarantor’s creditworthiness rather than the borrower’s in respect of the guaranteed portion. If the guarantor is the State and the other CRR requirements are met, this may, in theory, result in a risk weight close to zero and thereby a significant reduction in what would otherwise be a costly capital requirement.

In practice, however, the effect depends on how the guarantee is structured. A guarantee that in substance comes into play only upon an established final loss may have a materially weaker effect from a capital adequacy perspective than its economic risk-mitigating function might suggest.

The key questions are therefore whether the guarantee is sufficiently direct, whether payment may be demanded with sufficient predictability and whether the protection actually becomes available when the credit loss event occurs – rather than only after a protracted process to establish the final loss has been completed. The legal question is therefore not merely whether the State ultimately bears the risk, but when and on what terms the protection may in fact be called upon. In this respect, Boverket’s credit guarantee is no exception and must be assessed against the same criteria in order to determine what relief, if any, it affords in relation to capital requirements.

How Is Boverket’s Guarantee Structured?

From that perspective, Section 7 of Ordinance (2020:255) on a state credit guarantee for loans for residential construction is central, because that provision links compensation under Boverket’s credit guarantee to loss of principal upon a compulsory sale, another sale approved by Boverket, or a debt write-down approved by Boverket.

That structure suggests that the guarantee is, in practice, designed as protection against final loss, rather than as a payment guarantee that may be invoked already upon the debtor’s payment default. Functionally, Boverket’s credit guarantee therefore appears closer to an ordinary guarantee or ex post compensation than to a first-demand-type structure, which would more likely have provided the direct and effective protection capable of giving rise to capital adequacy relief under the CRR. This is also consistent with SOU 2017:108, in which the inquiry expressly proceeded on the basis that the credit guarantee, in its current form, is an ordinary guarantee.

The Swedish Financial Supervisory Authority’s consultation response to the inquiry points in the same direction (consultation response FI Dnr 18-6478, 2018-06-21). There, the Authority emphasised that the requirements of Article 215.1(a) – namely, that the guarantee must be enforceable in a sufficiently direct and effective manner in order to constitute eligible credit protection – must be satisfied, and therefore concluded that the guarantee has no effect on the capital requirements of credit institutions.

The issue, however, is not necessarily that payment is not made immediately. In an interpretative statement in its Q&A system (Q&A 2017_3576), the European Banking Authority (EBA) accepted that, under certain conditions, payment may be made in accordance with the original payment schedule. In other words, guarantee protection does not necessarily fall outside the CRR merely because the payout follows the loan’s original payment profile rather than being made immediately upon payment default. The more difficult question is whether a guarantee whose compensation presupposes realisation or an approved debt write-down provides the sufficiently direct and effective protection normally required for full capital adequacy relief under the CRR.

The question in relation to Boverket’s guarantee should therefore not primarily be framed as a simple yes-or-no question as to whether the guarantee “counts” under the CRR. The more precise question is whether the compensation mechanism in the individual case is sufficiently direct and predictable to permit the substitution approach, or whether that assessment can only be made after a closer analysis of the specific contractual terms. On the basis of the material currently available, there is no sound basis for describing the guarantee as an instrument that eliminates the lender’s capital cost. Rather, it appears uncertain, and in practice less likely, that the guarantee, even in an individual case, provides any meaningful capital adequacy relief. If any regulatory effect can be achieved at all, it appears, on the current material, to be limited and dependent on the detailed structure and operation of the arrangement in the individual case.

What Does This Mean in Practice for Banks?

The fact that the guarantee is unlikely to provide meaningful capital adequacy relief does not mean that it lacks value. It may still reduce economic credit risk, improve the bankability of projects and, in some cases, reduce the need for other forms of risk-bearing financing. Its principal function, however, appears to be credit risk mitigation in economic terms rather than a reliable instrument for regulatory capital relief.

From a banking perspective, it is precisely the interaction between economic risk and regulatory capital cost that is decisive. If economic risk is reduced but the capital requirement largely remains, the effect is materially smaller than a state guarantee might intuitively suggest. For banks, this means that Boverket’s credit guarantee, on the basis of the material currently available, can hardly be treated as an instrument that, in and of itself, resolves the capital cost arising from the lending.

That is also why banks and other lenders subject to capital adequacy rules attach considerable weight, in practice, to whether a guarantee or similar credit support has a clear and predictable effect within the capital adequacy framework.

It also explains why Boverket’s guarantee, notwithstanding its housing policy objective, appears less attractive to banks than a structure that could more clearly provide capital adequacy relief. For a bank that prices loans by reference to both credit risk and committed capital, that distinction is fundamental.

How Is the Guarantee Used in Practice?

Boverket’s annual reports for 2023–2025 show that use of the credit guarantee remains limited. The number of new guarantees amounted to 14 in 2023, 28 in 2024 and 20 in 2025. In recent years, the new guarantees have covered around 1,500–2,000 dwellings per year; in 2025, the figure was 1,613 dwellings. That corresponds to only a few per cent of annual residential construction, and an even smaller share of the residential construction that would be required. At a minimum, this is consistent with the view that the guarantee, notwithstanding its purpose, has not been as attractive to banks and other lenders as the need for new housing might otherwise justify.

Private Credit and the Regulatory Asymmetry

The way in which Boverket’s credit guarantee is structured is also of interest in light of the growing discussion around private credit. Banks operate within a capital requirements regime that directly affects pricing and capacity. Alternative lenders may, in many cases, operate under different regulatory conditions. If Boverket’s guarantee does not in practice reduce banks’ capital costs, it risks having limited effect precisely where regulatory friction is greatest.

At the same time, it cannot be assumed that all private credit actors fit readily within Boverket’s guarantee structure. In practice, the guarantee presupposes a lender that may be accepted by Boverket and that bears the risk in a manner that functions within the system. The broader point nevertheless remains: if banks do not derive a clear regulatory benefit from the guarantee while other forms of financing are less affected by capital requirements, an asymmetry arises that may affect which actors are in fact able or willing to finance residential construction.

Should the Guarantee Be Made More Effective?

At a time when too little is being built, and the State is already prepared to assume a measure of credit risk, there are strong reasons to ask whether the guarantee should be designed more effectively.

If the objective is also to facilitate bank financing, it is not enough for the guarantee merely to compensate for an established loss after the event. In that case, it should be structured so that the protection is more direct, more predictable and more usable from a capital adequacy perspective as well. This does not mean that the State must assume unlimited risk or accept a pure on demand model. The regulatory function should, however, carry greater weight in the design.

Reforms Under Discussion

State credit guarantees are, at the same time, already under review. During 2025, the Government circulated for consultation a memorandum concerning credit guarantees for loans for the provision of new housing in establishment and transition municipalities, that is, in substance, a new or specially subsidised guarantee model for certain housing projects.

So far as is known, no final legislative proposal has yet been presented. It is clear, however, that the guarantee system is under review. On the available material, the question of the guarantee’s function within the capital adequacy framework does not appear to have played a prominent role in that discussion. It would therefore be natural for that issue also to be taken into account when the State is, in any event, considering how future credit guarantees should be structured.

Concluding Remarks

Boverket’s credit guarantee may serve an important function by mitigating credit risk, enabling higher leverage and creating better conditions for projects that might otherwise struggle to attract financing. If the guarantee is also to become a more effective tool for bank financing, however, the distinction between economic risk-sharing and regulatory effect must be taken more seriously. At present, much suggests that the guarantee is better suited to covering final loss than to providing banks with clear capital adequacy relief. For that reason, there are good grounds for considering whether the guarantee can be made more attractive and more effective in a financing landscape in which the need for new housing is substantial and bank financing and private credit operate under different conditions.