A swift and decisive response of policy makers across OECD countries has helped businesses to bridge the short-term liquidity shortfalls due to the economic shock following the COVID-19 outbreak, avoiding immediate and widespread insolvency crises. Many countries have now entered a second phase of the pandemic and the shock is likely to translate into an enduring risk of a wave of corporate insolvencies as well as in a significant increase in leverage, depressing investment and job creation for long.

Building on OECD (2020a), which focused on the short-term risk of a liquidity crisis, this policy brief assesses two key types of risks in the medium and long-term:

  • Widespread distress and increase in leveraging. The number of non-financial corporations in distress, i.e. firms that are anticipated to have a negative book value of equity and therefore a high risk of insolvency, is expected to increase worldwide. At least two channels could be at work. First, the economic shock following the COVID-19 crisis diminished actual and expected sales and profits, thereby potentially putting downward pressure on the value of firms’ assets. Second, the injection of liquidity provided in the form of loan guarantees and new lines of credit increased firms’ leverage ratios and hence their default risk.1 To shed light on these challenges, we perform an accounting exercise in the spirit of Carletti et al. (2020) and Guerenni et al. (2020). Based on a sample of almost one million firms located in 14 European countries and in comparison to a business-as-usual counterfactual, we calculate the decline of net profit over a one year period, the associated decline in equity and the increase in leverage ratios.

  • The negative effect of debt overhang on investment. Higher levels of corporate debt require businesses to reduce investment in the aftermath of economic crises (Kalemli-Ozcan et al., 2019; Barbiero et al., 2020), thereby slowing down the speed of the recovery. Relying on regression analysis and looking at the historical relationship between investment and the financial leverage ratio at the firm level, we find that higher financial leverage tends to be associated with lower investment and that this negative relationship has been particularly pronounced during and after the global financial crisis (GFC). We then calculate the potential implications of this projected increase in leveraging as compared to normal times for investment ratios in the recovery period following the COVID-19 crisis.

Against these risks, the policy brief discusses options for policy makers to prevent widespread insolvencies and how to support firms without further increasing debt and leverage across firms. An additional layer of complexity concerns identifying the prevalence of non-viable firms, which could further undermine the strength of the recovery by locking-in resources in less productive firms (Adalet et al., 2018). The exceptional magnitude of the crisis and the high levels of uncertainty firms are still facing are likely to make the distinction between viable and non-viable firms more difficult. The risk of supporting potentially non-viable firms needs to be balanced against the risk of forcing viable and productive firms into premature liquidation.2 This is because insolvency frameworks tend to become less efficient in a time of crisis, especially when courts are congested, potentially leading to liquidation of a higher number of viable firms than desirable, with adverse effects on growth (Iverson, 2018).

To get around the necessity to identify non-viable firms at an early stage, it is critical to organise policy support under the premise of preserving optionality, i.e., helping firms weather the COVID crisis but regularly re-assessing their viability (e.g. stage-financing approach as suggested by Hanson et al., 2020). More broadly, one potential strategy for governments would be to adopt a multidimensional cascading approach.3 At first, policy makers could aim at “flattening the curve of insolvencies” by providing additional resources and restoring equity of distressed firms. Next, if those additional resources are not sufficient, they could encourage timely debt restructuring to allow distressed firms to continue operating smoothly. These two first steps are expected to reduce the number of viable firms that would be otherwise liquidated. Finally, to deal with firms that would still be non-viable despite public support and debt restructuring, governments could improve the efficiency of liquidation procedures to unlock potentially productive resources. Indeed, looking forward, policy makers will acquire new information on how the “post-pandemic” normal will look like and policy may need to facilitate the “necessary” reallocations implied by COVID-19.

Using a simple accounting exercise in the spirit of Carletti et al. (2020), we evaluate quantitatively the impact of the pandemic on firms’ long-term viability. The economic shock is modelled as a change in firms’ operating profits, resulting from the sharp reversal in sales and from firms’ limited ability to fully adjust their operating expenses. After calculating the decline in profits, also taking into consideration governments’ job support schemes implemented during the first phase of the crisis, the note sheds light on two different but related issues.4 First, it assesses the new hypothetical value of net equity (i.e. the difference between the book value of assets and liabilities) one year after the implementation of confinement measures. Firms whose net equity is predicted to be negative are classified in this framework as distressed, and thus at risk of being insolvent. This exercise informs about the amount of equity that would be needed to restore firms’ pre-crisis financial structure. Second, the note assesses the increase in firms’ leverage ratios caused by the reduction in equity relative to a “No-COVID” scenario. Box 2 provides the details of the accounting framework.

The analysis relies on financial statements of non-financial corporations from the latest vintage of the Orbis database. After applying cleaning procedures, the final sample consists of 872 648 unique firms, operating in both manufacturing and business non-financial services industries, for 14 relatively well-covered European countries.5 6 Importantly, as the objective of the exercise is to investigate the extent to which solvent firms may turn distressed due to the COVID-19 shock, we exclude from the sample firms that would have been distressed (e.g. firms with negative book value of equity at the end of 2018) and would have experienced negative profits even in normal times. It follows that the findings show an incremental – rather than total – effect following the COVID-19 shock. Moreover, firms in Orbis are on average disproportionately larger, older and more productive than in the population, even within each size class (Bajgar et al., 2020). As these firms are on average healthier than their smaller, younger and often less productive counterparts, the results from this analysis should be interpreted as a lower bound for the impact of the COVID-19 outbreak on firms’ solvency and leveraging.

The magnitude of the sales shock during confinement months is based on the first-round demand and supply shocks computed at a detailed sectoral level by del Rio-Chanona et al. (2020).9 To quantify the supply shock, they classified industries as either essential or non-essential and constructed a Remote Labour Index, which measures the ability of different occupations to work from home: the supply shock is not binding for essential industries, while inversely proportional to the capacity to telework for non-essential ones. To quantify the demand shock, del Rio-Chanona et al. drew on a study of the potential impact of a severe influenza epidemic developed by the US Congressional Budget Office. In this note, we identify the resulting sector-specific – but country invariant – shock as the largest between the supply and the demand shock.10

The model presents two alternative scenarios with respect to the duration of the shock (see Table A.1 in the Annex):

  • An upside scenario, which foresees a sharp drop in activity lasting two months (the confinement period), followed by progressive but not complete recovery in the remaining part of the year. Consistent with the OECD’s June 2020 Economic Outlook projections, activity levels are expected to still fall short of pre-crisis levels one year after the start of the implementation of confinement measures. The recovery path is dependent on the initial shock, so that the most severely hit sectors face a larger absolute decline in revenues also after confinement, but the speed of the recovery is assumed, for simplicity, to be the same across sectors.

  • A downside scenario, which overlaps with the upside scenario for the first seven months, but then embeds a second, relatively smaller, outbreak from the eighth month onwards, accompanied by more limited lockdowns.11 12

The current developments of the pandemic, characterised by localised outbreaks and uncertainty on the strength of virus resurgence suggest that the recession may be more deeper than modelled in the upside scenario but not as severe as in the downside one. It follows that the two scenarios could be more generally interpreted as a lower and an upper bound with respect to the magnitude of the shock.

The estimated decline in profits is sizeable, on average between 40% and 50% of normal time profits (depending on the scenario considered).13 Following this sharp reduction, 7.3% of otherwise viable companies would become distressed in the upside scenario and 9.1% in the downside scenario (Figure 1, top left panel).14 The highlighted incremental effect following the COVID-19 shock implies that the total number of distressed firms would double compare to “normal times”, as we estimate approximately 8% of firms being endangered in a No-COVID scenario.

The results are quite heterogeneous across sectors and type of firms. The percentage of otherwise viable companies becoming distressed in the “Accommodation and food service activities” sector reaches 26% and 32% in the upside and the downside scenarios respectively, while it is almost null in the “Information and communication” and “Professional services” ones (Figure 1, top right panel). The “Transports”, “Wholesale and retail trade”, as well as “Arts, entertainment and recreation” and “Other services activities” sectors are also severely hit by the crisis. The percentage of distressed firms in manufacturing is instead below average.

More broadly, and consistent with the diverse ability to rely on innovative technologies and teleworking arrangements, tangible-intensive sectors are relatively more affected than intangible-intensive ones (Figure 1, middle left panel). Similarly, more productive companies are relatively less impacted than low-productivity firms; yet, the estimated percentage of firms in the top quartile of the productivity distribution becoming distressed is not negligible (5.4% and 6.8% in the upside and the downside scenarios, respectively; middle right panel). In addition, old and large firms are better positioned to face the shock compared to their younger and smaller counterparts (Figure 1, bottom panels).

The reduction in equity relative to a business-as-usual scenario has immediate consequences on firms’ leverage ratios: the ratio of total liabilities to total assets would increase by 6.7 percentage points in the upside scenario and 8 percentage points in the downside scenario for the median firm in the sample (Figure 2, left panel). Importantly, while leverage ratios are estimated to substantially increase due to the COVID-19 shock over the whole range of the pre-crisis distribution of leverage, the new distribution of firms according to their leverage ratio shows a larger portion of firms with very high leverage ratios, underlying the likelihood of large-scale over-indebtedness (Figure 2, right panel).

Similarly, the sizeable decline in profits relative to business-as-usual may impair firms’ ability to service their debt. The left panel of Figure 3 shows that, despite assuming no increase in interest payments compared to normal times, 30% of the companies are not profitable enough to cover their interest expenses in the upside scenario (36% in the downside scenario) – i.e., they have an interest coverage ratio lower than unity. In line with this, the interest coverage ratio is estimated to be approximately halved due the COVID-19 outbreak for the median firm in the sample. The right panel of Figure 3 disaggregates results at the sector level, showing once again large heterogeneity across sectors and pointing out that a consistent portion of firms in the “Accommodation and food service activities”, “Arts, entertainment and recreation” and “Transport” sectors will find it difficult to service their debt. Unsurprisingly, young, small and less productive companies are predicted to be hit more severely by the crisis also according to this metric.

The increase in the level of indebtedness and risk of default can push firms towards the so-called “debt overhang” risk. When a firm has a high outstanding debt on which the likelihood of default is significant, the return from any investment tends to disproportionately benefit senior debt-holders compared to shareholders (Myers, 1977).15 Therefore, the expected return from investment needs to be high enough to cover the value of outstanding debt held by senior creditors and to offer an additional return to new investors and equity holders – e.g. having a positive net present value may not suffice for an investment to be economically favourable (Chatterjee, 2013).16 Ultimately, the reduced incentives to invest and limited access to new credit for highly indebted companies generates pressure to deleverage by cutting costs and downsizing, even in companies with profitable investment opportunities, potentially slowing down the recovery from the current crisis. Recent financial indicators corroborate that the combination of negative pressure on sales, high uncertainty about future sales and profits, and growing debt burdens has increased the risk of default, putting particular downward pressure on corporate credit ratings.17 In March 2020, 389 non-financial corporations across OECD countries saw a decrease in their credit rating, compared to 61 downgrades in March 2019.18

In order to more formally assess how the rising tide of debt associated with the COVID-19 outbreak would affect investment and to size the potential magnitude of the effect, we investigate empirically the historical relationship between indebtedness and investment, as well as the specific features characterising the relation during the global financial crisis (GFC).19 Box 3 provides the details of the methodological framework.

Consistent with previous studies (e.g. Kalemli-Ozcan et al., 2019; Barbiero et al., 2020), the results from the panel model confirms the existence of a negative and statistically significant relationship between financial leverage and investment. Interpreting results causally, the magnitude of the effect is not negligible. A 19 p.p. (43 p.p.) increase (decrease) in the debt to assets ratio (interest coverage ratio) – e.g. approximately equivalent to a one-standard deviation in our sample – is expected to decrease the investment ratio by 6 p.p. (1 p.p.). In other words, such a change in financial leverage (the interest coverage ratio) would explain 15% (3%) of the observed variation in investment over the period.20 Results are confirmed when focusing exclusively on tangible or intangible investments, and are consistent with respect to a wide range of robustness checks – for instance, looking separately at the pre- and post-GFC periods or using quartiles of the financial leverage ratio variable, rather than its continuous value, to capture firms’ indebtedness levels.

Figure 4 further explores the implied magnitude of the estimated relationship between investment and financial leverage ratios by assuming an increase in the debt to total assets ratio comparable to that predicted by our accounting model in the post-COVID period.21 Specifically, Figure 4 illustrates the inferred decrease in investment to fixed assets under the hypothetical increase in financial leverage as shown in Figure 2 (left panel) for the median firm in the sample. Investments to fixed assets would decrease by 2 percentage points in the upside scenario and 2.3 in the downside one.

The outcome of the cross-sectional model strengthens these findings and shows that the relationship holds in the presence of a large shock such as the GFC (see Table A.4). An increase in firms’ financial leverage has on average a significant and negative impact on the growth of investment in the aftermath of the crisis. Similarly, the positive coefficient for the interest coverage ratio implies that a decrease in the ability to service debt is also associated with lower investment growth. However, the effect of a change in debt on investment is heterogeneous across firms and is increasing with initial leverage levels (Figure 5): firms that entered the GFC with a higher financial leverage ratio experienced a sharper decline in investment; on the contrary, the relation can turn positive, and thus an increase in debt could foster investment, for firms with very low initial indebtedness levels. The effect is economically considerable. For instance, a 15 percentage points increase in the change in financial leverage – approximately equivalent to a one-standard deviation – implies a decrease in the investment rate equivalent to 18% of its average change for firms with a financial leverage ratio around 0.3, while it implies a 5% increase for firms with no previous debt.

Overall, these results suggest that debt-overhang could hamper investment and impede a fast recovery following the COVID-19 outbreak, given the record-high debt levels at the beginning of 2020 and the ongoing and expected rise in corporate debt to face the economic consequences of the pandemic. Moreover, the analysis hints that looking exclusively at average effects may convey policy makers only a partial picture and that a further increase in debt may not be the appropriate answer for highly indebted firms, both from a viability and an investment perspective.

The empirical analysis stresses that the rise of corporate debt could threaten the recovery, suggesting that governments should be careful when designing support packages. In the initial phase of the COVID-19 crisis, temporary deferral or repayments of loans either by private agents (e.g. banks in the Netherlands) or public sources (e.g. loans by the Ministry of Tourism in Spain) played a key role to relieve financially distressed businesses and prevent early insolvency. Loan guarantees also help distressed firms to meet their immediate financial commitments, avoiding widespread defaults (e.g. the Überbrückungskredite – loan guarantees for short-term credits – processed by specialised Austrian banks). However, such support may not address the issue of their long-term viability due to the associated rise in indebtedness. The rest of the policy brief sheds light on various policy options to support distressed firms while not compromising their ability to invest. First, it focuses explicitly on the design of crisis-related measures and on the necessity to favour equity-type of financing over debt to recapitalise distressed firms. Second, it investigates the potential role of debt resolution mechanisms in mitigating debt overhang and in sorting-out viable and non-viable firms.

Increasing equity capital provides a way to support viable businesses without raising corporate debt. Relative to increases in debt, additional equity improves leverage ratios and reduces interest coverage ratios, thereby reducing corporate refinancing costs and benefitting a potential recovery. In times of high uncertainty over future sales growth, equity financing may also be desirable from the viewpoint of entrepreneurs, given that equity acts like an automatic stabiliser.

Governments have various policy options to leverage on equity financing to support viable businesses. On the one hand, authorities can directly provide equity capital through public equity injections in firms (e.g. Alitalia and Lufthansa by the Italian and German governments). Indirect options, such as convertible bonds or state guaranteed loans against higher profit tax rates in the future, are also possible. On the other hand, policymakers can incentivise the uptake and provision of equity capital from private investors to firms through structural reforms, e.g. of corporate tax systems, and acting as a catalyst of private resources by developing equity co-funds. An overview of currently existing equity programmes across OECD countries is provided in OECD (2020h; Table 1 on page 21).

Equity injections can help firms, which suffer from financial difficulties solely due to COVID-19 but are likely to return to profitability afterwards, to raise much-needed cash to finance their working capital while keeping assets free for raising debt in the future. Authorities would, however, need to ensure that losses for taxpayers are minimised and that the government does not intervene in management. Moreover, it is important to ensure that such support is state-contingent and includes mechanisms to incentivise all parties to wind down support when economic conditions improve (OECD, 2020c; OECD, 2020d).

Hybrid instruments like preferred equity appear well-adapted (OECD, 2020h). Preferred equity provides a senior claim to dividends and assets in case of liquidation as compared to common equity. Consequently, the owners of common shares, i.e. the owners with voting rights, have an incentive to manage the company relatively prudently, given that they only receive a dividend once the dividend payments on preferred shares are paid. Since preferred shares entail no voting rights, companies can raise funds without diluting control. Moreover, taxpayer could potentially profit from these investments: if the share of firms facing losses in the medium and long-term remains below a certain threshold, public equity injections may even lead to a higher return on investments than compared to debt injections (OECD, 2020h). Temporary forms of preferred equity, e.g. retractable preferred equity, would help to formulate an exit strategy in advance, thus further reducing risks for taxpayers.

Public equity injections come not without disadvantages. Given that preferred equity is subordinate to debt, equity injections can entail potentially larger losses than debt. The magnitude of these losses depends on whether the government can realise a positive return on investment and whether the number of firms in high distress remain below a certain threshold (OECD, 2020h). Furthermore, equity injections require agreement on the market value of equity and the required return for investors. Retractable preferred equity further needs agreement on maturity and exit conditions. While valuation and pricing may be much less of a problem for relatively large companies, smaller companies with no shareholders other than the owner are difficult to value. The government could team up with private investors, e.g. private equity and venture capital investors, to ensure a quick and efficient valuation. Valuation at the sector-level may help as well. Crucially, governments need to ensure that competition on markets is not overly distorted and that equity injections do not crowd out other investors. In order to avoid such risks, international coordination, for instance within the EU, is key in designing rescue packages for hardly hit cross-country industries (e.g. airlines).

Supporting SMEs and start-ups financing needs may require a different and more comprehensive approach, as equity markets for small and medium-sized, as well as young, firms are thinner and often lacking altogether.22 This makes the valuation of equity capital and thus the design of the injection more difficult. Besides direct equity injections, policymakers could revert to more indirect measures. For instance, according to the Business Continuity Insurance framework proposed by Hanson et al. (2020), repayment could be linked to businesses’ returns: firms that recover most robustly would pay back more, in the form of future taxes, while those that struggle longer would pay back less. Such support would have several advantages. It could contribute to flattening the curve of bankruptcies while limiting the risk of costly defaults. In addition, agreements to pay higher taxes in the future against guaranteed credits would be easier to monitor for authorities, than a potentially large number of equity injections in a large number of single entities.

While subject to the existence of sufficient fiscal space, another useful measure to address SME funding needs without raising debt consists in converting government (crisis related) loans into grants. For instance, in the United States, loans obtained through the “Paycheck Protection Program” could be turned into grants conditional on the firm spending at least three-quarters of the money on payroll expenses and the rest on rent and utility bills. Similarly, the German government launched the “Immediate Assistance Programme” (Soforthilfeprogramm) to provide grants to small businesses, self-employed individuals and freelancers, under the conditions of using the funds to mainly cover rental and leasing expenses; applications should be filed directly with the government of the Land of residence, and the maximum amount of the grant is set proportional to firms’ size.

Debt-equity swaps constitute another tool to address high leverage. They involve the conversion of outstanding debt that cannot be repaid into equity of an otherwise viable company. Debt-equity swaps may appear attractive in theory, but raise some implementation issues. A debt-equity conversion requires the estimation of the market value of debt and equity and an agreement between shareholders and the debtholder on an exchange ratio. The lack of equity markets for SMEs, in particular smaller ones, impedes a cost-efficient estimation of the market value of equity. Consequently, debt-equity swaps appear more appropriate to address elevated leverage in circumstances where agreements on underlying terms are more likely to be reached, e.g. subsidiaries of a larger firm, than as a more general policy tool.

Governments have additional policy tools to stimulate the uptake and provision of equity capital. One way for policymakers to leverage on the need for equity in the post-COVID-19 world would be to grant an allowance for corporate equity (ACE). Such an allowance would partially or totally offset the tax benefits of using debt financing and make equity financing more attractive. Their design should however ensure that multinationals do not exploit ACE for tax-planning and that their fiscal cost is acceptable, for instance by granting them to new equity capital only. In the OECD area, a few countries (such as Italy and Belgium) have already introduced ACE or experimented with it in the past and their experience can serve as example (Zangari et al., 2014; Hebous and Ruf, 2017).23 Moreover, deductions on income taxes and reliefs on the taxation of capital gains for eligible investments can foster the provision of private equity capital. Such tax incentives are often used to stimulate investments in high-risk, early-stage businesses, e.g. as in the UK’s Enterprise Investment and Seed Enterprise Investment scheme, but could potentially be extended also to a wider set of firms, e.g. smaller companies facing tight financing frictions.

Besides immediate short-term measures aimed at dealing with the economic consequences of COVID-19, there are options to ensure that equity markets continue to develop, including by widening access to equity markets for smaller firms, e.g. through reducing costs and streamlining listing requirements (Wehinger, 2014). For instance, COVID related equity programmes could speed up the implementation of the Capital Market Union in European countries, which in turn could help to address intra-European segmentation along national boundaries. Similarly, policy makers can improve the development and attractiveness of equity markets by using financial literacy as a tool to boost stock market participation and financial knowledge of entrepreneurs.

Equity and quasi-equity injections might prove insufficient to allow firms to operate normally if leverage ratios and risk of default remain high. For those firms, reducing the debt burden through debt restructuring can change both the timing of the potential default and their possibility to invest (Frantz and Instefjord, 2019). Most countries have already modified their insolvency framework to give insolvent firms a chance to survive in the short-run, for instance by relaxing the obligation for directors to file for bankruptcy once insolvent (e.g. France, Germany, Luxembourg, Portugal and Spain) or by relaxing creditors’ right to initiate insolvency proceedings as done in Italy, Spain, Switzerland and Turkey (OECD, 2020e; INSOL International-World Bank Group, 2020). Examples of recent temporary insolvency measures in response to COVID-19 are presented in Box 4.

However, more structural changes to features of insolvency regimes, which can be a barrier to successful restructuring, could help to co-ordinate creditors’ claims in a manner that is consistent with preserving the viability of the firm. This crisis can provide an opportunity for reforms which are often hard to implement and take time to design.24 Policy actions in these areas could rely on several pillars. First, establishing legal conditions that would increase incentives for investors to provide new financing to financially distressed firms. Second, favouring pre-insolvency procedures that are associated with a higher rate of success. Third, establishing specific procedures for SMEs, including promoting informal debt restructuring as they run a higher risk to be liquidated in formal insolvency processes. Fourth, establishing specific out-of-court procedures that have proven to be effective in time of systemic crisis. Finally, ensuring an efficient liquidation process and providing the institutional conditions for a fresh start to unlock productive resources is key.

Continuity of firm operations during restructuring increases the chances of a successful restructuring but often requires firms to have access to bridge financing. However, access to new funds may be difficult when the debt levels are already high and the risk of default is significant, leading to debt overhang. Across the OECD, new financing can have either no priority at all over existing creditors or priority over only unsecured creditors or over both secured and unsecured creditors. In normal times, insolvency regimes have to balance incentives for debtors to invest and take risks with incentives for creditors to supply funds. Therefore, new financing should be granted priority ahead of unsecured creditors but not over existing secured creditors since this would adversely affect the long-term availability of credit and legal certainty (Adalet McGowan and Andrews, 2018). Yet, several OECD countries currently do not offer any priority to new financing, so granting it over unsecured creditors would be beneficial. Further, in the context of the current crisis and assuming that the extensive guarantees and liquidity injections reach the right firms, the blocking of the “credit channel” might not be the main concern. An alternative but more controversial option to improve access to new financing is to temporarily suspend also the priority enjoyed by secured creditors in favour of new investors when they invest in distressed firms (Gurrea-Martínez, 2020).

Efficient pre-insolvency frameworks and debt restructuring could help to address debt overhang by lowering the negative impact of deleveraging on GDP growth and quickening the resolution of non-performing loans (Carcea et al., 2015; Bricongne et al., 2017). While a majority of OECD countries has some type of pre-insolvency legislation, until recently they were generally missing in non-European OECD countries (Adalet McGowan and Andrews, 2018). A number of countries has strengthened out-of-court procedures in recent years. For example, in 2018, Belgium, granted the courts the ability to endorse a settlement between a debtor and two or more of its creditors to make it enforceable. Lithuania overhauled the insolvency regime in 2020, accelerating timely initiation and resolution of personal and corporate insolvency proceedings and increasing returns for creditors, bringing them among the countries with the most efficient insolvency regimes according to the OECD indicator (OECD, forthcoming).25 In addition, several countries have encouraged lenders to reach out-of-court agreements with debtors materially affected by COVID-19, especially when these agreements just involve a deferral of loan repayments (Australia, China, India, Malaysia, and Singapore). More generally, introducing preventative restructuring or pre-insolvency frameworks, for instance as in the EU Directive on Preventive Restructuring Frameworks and Second Chance, could be accompanied by other incentives for private creditors to restructure debt, such as tax incentives (e.g. tax exemption for creditors who forgive part of debt). Effective design of such policies can be based on existing guidelines, such as the World Bank’s Toolkit for Out-of-Court Restructuring (World Bank, 2016).

Small and medium-sized enterprises (SMEs) may warrant different treatment from other firms in a debt restructuring strategy, as complex, lengthy and rigid procedures, required expertise and high costs of insolvency can be demanding for this category of firms. Indeed, SMEs are more likely to be liquidated than restructured, since they have to bear costs that are disproportionately higher than those faced by larger enterprises. For instance, the costs of bankruptcy for small and medium-sized businesses in the U.S. are often estimated at 30% of the value of the business and two-thirds of them are liquidated rather than reorganised (Skeel, 2020). In the current juncture with a high risk of insolvency among SMEs, the social cost of an inefficient debt restructuring for SMEs could be very large.

Against this background, formal procedures can be simplified for SMEs (see Box 5) and informal procedures, which typically avoid the procedural complexities and timelines of court proceedings and are often associated with better outcomes for SMEs, can be adopted relatively quickly (World Bank, 2020). In that respect, there are three major challenges specific to SMEs. First, while negotiation requires the participation and leadership of an honest broker, public authorities might not be the best informed to take decision on very small companies as they lack the detailed information and the administrative capacity to use this information. Second, as the amount of loans is often small and bound by different covenants, it is more difficult to pool and securitise them. Third, reaching an agreement among different creditors might be difficult, leading to a higher probability to liquidate the firm instead of restructuring.

To provide incentives for private creditors to restructure SMEs debt, Blanchard et al. (2020) suggest for instance the following ad-hoc process. If the different private creditors reach an agreement and agree on debt restructuring, the government would accept a haircut equivalent to private creditors of the same rank and offer a fixed continuation premium. This continuation premium would provide incentives for banks to make the right social decision and addresses the difference between the social and the private value of the firm. By contrast, if creditors do not reach a consensus and liquidate the firm, the government would take its full claim on its rights.

In-court debt restructuring for large firms appears broadly efficient in normal times, but during systemic crises case-by-case restructuring can become difficult, private capital is limited and co-ordination problems become more serious. In these conditions, court-supervised restructuring can be too time-consuming. Against this background, government agencies could prioritise out-of-court renegotiations whenever possible, a strategy that proved to be successful after the global financial crisis (Bernstein, Lerner, and Mezzanotti, 2019; Hotchkiss, Smith, and Strömberg, 2012). When out-of-court restructuring is difficult due to too many creditors, a centralised out-of-court approach might be desirable (Box 6).

Providing equity support for distressed firms and ensuring debt restructuring should reduce a build-up of undesirable bankruptcies, but some firms will still remain unviable in the post-COVID world (e.g. due to their business model, their financial situation or their product specialisation). Against this risk, policy makers need to address several challenges to ensure that the liquidation process of those firms is efficient and provide the institutional conditions for a fresh start by removing barriers that might push debtors to delay liquidation, easing the reallocation that the potentially permanent shifts in the nature of economic activity might induce in the longer term.26

Ensuring the highest possible recovery rate for creditors. When the number of distressed firms is too large, the courts become overwhelmed, standard insolvency procedures work less effectively, and the recovery rates for creditors can be reduced, potentially at fire-sale prices.27 Any reforms that can simplify and speed up in-court processes would help in this respect. In the short term, a strategy aiming at flattening the curve of insolvencies combined with increased resources for the court system, for instance by adding new temporary judges on insolvency procedures or reallocating judges depending on the busiest jurisdictions, would contribute to increase the recovery rate of creditors.

Ensuring that liquidation is established by an independent broker. Public agencies such as public development banks in charge of loan guarantees may not be well placed to negotiate liquidation – with their own balance sheets exposed, they may be inclined to “extend-and-pretend” distortions in their actions (Bertay, Demirguc-Kunt and Huizinga, 2015).Therefore, one challenge for policy makers is to establish an independent organisation to ensure that decisions with respect to liquidation and debt restructuring are not distorted (Hege, 2020).

Reducing specific barriers to market exit for small firms. The corporate versus personal distinction in assets and liabilities is often blurred for small firms, either because lenders require personal guarantees or security – e.g. a second mortgage on the owner’s home – or because prior to incorporating and obtaining limited liability protection, entrepreneurs typically use personal finances (Berkowitz and White, 2004; Cumming, 2012). As a result, corporate insolvency may lead to personal insolvency once a business fails, even where the business is a separate legal entity. In that context, the type of personal insolvency regime matters to reduce the scars from the crisis, in particular by enabling a post-insolvency second chance for entrepreneurs and the availability of a “fresh start” – i.e. the exemption of future earnings from obligations to repay past debt due to liquidation bankruptcy. This could reduce the costs and the stigma of failure associated with insolvency, which is one of the commonly cited barriers to entrepreneurship and to firm dynamism. 28 Despite the importance of personal insolvency regimes, previous OECD research has shown that between 2010 and 2016, reform in this area was very limited, only five countries in the OECD (Chile, Greece, Lithuania, Latvia and Spain) lowered personal costs to failed entrepreneurs (Adalet McGowan and Andrews, 2018). Many European countries are already lowering time to discharge to 3 years to be in line with the EU Directive on Insolvency and Second Chance (e.g. Germany), but they could try to expedite this part of the reform, which can facilitate reallocation (e.g. Spain is considering this option).

To conclude, Figure 6 summarizes the main challenges for policy makers and the surveyed instruments to address them.


Acharya, V., D. Gale, and T. Yorulmazer, (2011), “Rollover risk and market freezes,” The Journal of Finance, 66 (4): 1177-1209.

Adalet McGowan, M., and D. Andrews, (2016), “Insolvency regimes and productivity growth: A framework for analysis”, OECD Economics Department Working Papers, No. 1309, OECD Publishing, Paris.

Adalet McGowan, M., and D. Andrews, (2018), “Design of insolvency regimes across countries”, OECD Economics Department Working Papers, No. 1504, OECD Publishing, Paris.

Adalet McGowan, M., D. Andrews and V. Millot, (2017), “The walking dead? Zombie firms and productivity performance in OECD countries”, OECD Economics Department Working Papers, No. 1372, OECD Publishing, Paris.

Adalet McGowan, M., D. Andrews and V. Millot (2017), “Insolvency regimes, zombie firms and capital reallocation”, OECD Economics Department Working Papers, No. 1399, OECD Publishing, Paris.

Allen & Overy, (2020), Covid-19 coronavirus measures impacting insolvency proceedings and enforcement.

Almeida, H., I. Cunha, M.A. Ferreira and F. Restrepo, (2017), “The real effects of credit ratings: The sovereign ceiling channel”, The Journal of Finance, Vol. 72(1): 249-290.

Bajgar, M., Berlingieri, G., Calligaris, S., Criscuolo, C., and J. Timmis, (2020), “To use or not to use (and how to use): coverage and representativeness of Orbis data”, OECD Science, Technology and Industry Working Papers, No. 2020/06, OECD Publishing, Paris.

Barberis, N., and R. Thaler, (2003), “A survey of behavioral finance”, Handbook of the Economics of Finance, vol.1.

Barbiero, F., A. Popov and W. Marcin, (2020), “Debt overhang, global growth opportunities, and investment”, Journal of Banking and Finance, Vol. 120, Article 105950.

Becker, B., and V. Ivashina, (2014), “Cyclicality of credit supply: Firm level evidence”, Journal of Monetary Economics, Vol. 62, pp. 76-93.

Becker, B., U. Hege and P. Mella-Barral, (2020), “Corporate debt burdens threaten economic recovery after COVID-19: Planning for debt restructuring should start now”, VoxEU.org, 21 March.

Berkowitz, J., and M. White, (2004), “Bankruptcy and small firms’ access to credit”, RAND Journal of Economics, Vol. 35(1): 69-84.

Bernstein, S., J. Lerner, and F. Mezzanotti, (2019). “Private equity and financial fragility during the crisis”, Review of Financial Studies, 32(4)

Bertay, A., A. Demirguc-Kunt and H. Huizinga, (2015), “Bank ownership and credit over the business cycle: Is lending by state banks less procyclical?”, Journal of Banking and Finance, Vol. 50(3): 326-339.

Bird and Bird, (2020), COVID-19: International Restructuring & Insolvency update.

Blanchard, O., T. Philippon and J. Pisani-Ferry (2020), “A new policy toolkit is needed as countries exit COVID-19 lockdowns”, Peterson Institute for International Economics Policy Brief, N. 20-8.

Boot, A., E. Carletti, H. H. Kotz, J. P. Kranhen, L. Pelizzon and M. G. Subrahmanyam, (2020), “Coronavirus and financial stability 3.0: Try equity-risk sharing for companies, large and small”, VoxEU.org, 2 April.

Boschmans, K., and L. Pissareva, (2017),”Fostering markets for SME finance”, OECD SME and Entrepreneurship Papers, No. 6.

Branzoli, N., and A. Caiumi, (2018),”How effective is an incremental ACE in addressing the debt bias? Evidence from corporate tax returns”, European Commission Taxation Papers, No. 72 - 2018.

Bricongne, J.C, M. Demertzis, P. Pontuch and A. Turrini, (2016). “Macroeconomic relevance of insolvency frameworks in a high-debt context: An EU perspective”, European Economy - Discussion Papers, No.2015- 032, Directorate General Economic and Financial Affairs, European Commission.

Brunnermeier, M., and A. Krishnamurty, (2020), “COVID-19 SME evergreening proposal: Inverted economics”, mimeo.

Caballero, R., T. Hoshi and A. Kashyap, (2008), “Zombie Lending and Depressed Restructuring in Japan”, American Economic Review, Vol. 98(5): 1943-1977.

Campenhout, G., T. V. Caneghem, (2013),”How did the notional interest deduction affect Belgian SME’s capital structure?”, Small business economics, Vol. 40(2).

Carcea, M., D. Ciriaci, C. Cuerpo, D. Lorenzani and P. Pontuch, (2015), “The Economic Impact of Rescue and Recovery Frameworks in the EU”, EU Discussion Papers, No. 004.

Carletti, E., T. Oliviero, M. Pagano, L. Pelizzon and M. G. Subrahmanyam, (2020), “The COVID-19 shock and equity shortfall: Firm-level evidence from Italy”, CEPR Discussion Paper No. 14831, also published COVID Economics: Vetted and Real-Time Papers , issue 25.

Chatterjee, S., (2013), “Debt overhang: Why recovery from a financial crisis can be slow”, Business Review, Federal Reserve Bank of Philadelphia, issue Q2, p.p. 1-9.

Cumming, D., (2012), “Measuring the effect of bankruptcy laws on entrepreneurship across countries”, Journal of Entrepreneurial Finance, Vol. 16(1): 80-86.

Dell’Ariccia, G., D. Igan, L. Laeven and H. Tong, (2016). “Credit Booms and macrofinancial stability”, Economic Policy 31(86): 299-355.

del Rio-Chanona, R. M., P. Mealy, A. Pichler, F. Lafond and J. D. Farmer, (2020), “Supply and demand shocks in the COVID-19 pandemic: An industry and occupation perspective”, COVID Economics: Vetted and Real-Time Papers, issue 6.

Demmou, L., I. Stefanescu and A. Arquié, (2019), “Productivity growth and finance: the role of intangible assets - a sector level analysis”, OECD Economics Department Working Papers, No 1547, OECD Publishing.

Dentons, (2020), Changes in Bankruptcy Law: An overview of (temporary) measures in response to COVID-19.

European Commission, (2019), EU Directive on Preventive Restructuring Frameworks and Second Chance.

Frantz, P., and N. Instefjord, (2019), “Debt overhang and non-distressed debt restructuring”, Journal of Financial Intermediation, Vol. 37 (C): 75-88.

Frühwirth, M., and M. Kobialka, (2011), “Do Equity Tax Shields Reduce Leverage? The Austrian Case”Available at SSRN: https://ssrn.com/abstract=1458245.

Gal, P., (2013), “Measuring total factor productivity at the firm level using OECD-ORBIS”, OECD Economics Department Working Papers, No. 1049, OECD Publishing, Paris.

Gurrea-Martinez, A., (2020), “Insolvency law in times of COVID-19”, Oxford Law Faculty Blog.

Hanson, S., J. Stein, A. Sunderam and E. Zwick, (2020), “Business Continuity Insurance: Keeping America’s lights on during the pandemic”, White Paper of Becker Friedman Institute.

Hebous, S., and M. Ruf, (2017),”Evaluating the effects of ACE systems on multinational debt financing and investment”, Journal of Public Economics, Vol. 156: 131-149.

Hege, U., (2020), “Corporate debt threatens to derail recovery”, TSE Mag #20.

Hotchkiss, E.S., P. Stromberg and D. Smith, (2012), “Private equity and the resolution of financial distress”, AFA 2012 Chicago Meetings Paper, ECGI - Finance Working Paper, No. 331.

Iverson, B., (2018), “Get in line: Chapter 11 restructuring in crowded bankruptcy courts”, Management Science, Vol. 64 (11): 5370-5394.

INSOL International-World Bank Group, (2020), Global Guide Corporate Insolvency: Responses in Times of Covid-191: Report,

Jordà, O., M. Schularick and A.M. Taylor, (2013), “When credit bites back”, Journal of Money, Credit and Banking, Vol 45: 3-28.

Kalemli-Ozcan, S., L.A. Laeven, and D. Moreno, (2019) “Debt Overhang, (2019) Rollover Risk, and Corporate Investment: Evidence from the European Crisis”, ECB Working Paper, No. 2241

Kent, P., (1993), “The London Approach”, Bank of England Quarterly Bulletin.

Guerini, M., L. Nesta, X. Ragot and S. Schiavo, (2020), “Firm liquidity and solvency under the COVID-19 lockdown in France”, OFCE policy brief, No.76.

Miller, M., and J. Stiglitz, (2010), “Leverage and asset bubbles: Averting armageddon with Chapter 11?”, NBER Working Papers, No. 15817.

Murray, M., (2020), “Managing the Insolvency Curve in Australia, Oxford Business Law Blog.

Myers, S. C., (1977) “Determinants of Corporate Borrowing”, Journal of Financial Economics, Vol. 5 (2), 147–175.

OECD, (2020a), “Corporate sector vulnerabilities during the COVID-19 outbreak: Assessment and policy responses”, Tackling Coronavirus Series.

OECD, (2020b), Initial impact of COVID-19 pandemic on the non-financial corporate sector and corporate finance", Forthcoming Tackling Coronavirus Series.

OECD, (2020c), “The COVID-19 crisis and state ownership in the economy: Issues and policy considerations”, Tackling Coronavirus Series.

OECD, (2020d), “Supporting businesses in financial distress to avoid insolvency during the COVID-19 crisis”, Tackling Coronavirus Series.

OECD, (2020e), “National corporate governance related initiatives during the COVID-19 crisis: A survey of 37 jurisdictions”, Tackling Coronavirus Series.

OECD, (2020f), “Corporate bond market trends, emerging risks and monetary policy”, OECD Capital Market Series.

OECD, (2020g), “Start-ups in the time of COVID-19: facing the challenges, seizing the opportunities”, Tackling Coronavirus Series.

OECD (2020h),”COVID-19 Government Financing Support Programmes for Businesses”, OECD Paris.

Ozdamar, M., B. Tanyeri and L. Akdeniz, (2020),”The effect of tax regulation on firm value: the Turkish case of Allowance for Corporate Equity (ACE) regulation”, Applied Economics Letters.

Panier, F., Pérez-González, F., and P. Villanueva, (2013), “Capital Structure and Tax: What Happens When You (Also) Subsidize Equity”, mimeo.

Princen, S., (2012),”Taxes do affect corporate financing decisions: The case of Belgian ACE”, CESifo working paper series, No. 3713.

Skeel, D., (2020), “Bankruptcy and the coronavirus”, Brookings Economic Studies.

Squire Patton Boggs, (2020), Impact of COVID-19 on Insolvency Laws: How Countries Are Revamping Their Insolvency and Restructuring Laws to Combat COVID-19.

van Rooij, M., A. Lusardi and R. Alessie, (2011), “Financial literacy and stock market participation”, Journal of Financial Economics, Vol. 101(2): 449-472.

Vattermoli, D., (2020), “Pandemic and Insolvency Law: the Italian Answer, Oxford Law Blog.

Woo, D., (2000), “Two approaches to resolving nonperforming assets during financial crises”, IMF Working Paper, No. 33.

Wooldridge, J. M., (2009), “On estimating firm-level production functions using proxy variables to control for unobservables”, Economics Letters, 104: 112–114.

World Bank, (2016), “Principles for effective insolvency and creditor/debtor regimes”, World Bank Group.

World Bank, (2017), “Report on the treatment of MSME insolvency”, World Bank Group.

World Bank, (2018), “Saving entrepreneurs, saving enterprises: Proposals on the treatment of MSME insolvency”, World Bank Group.

Zangari, E., (2014), “Addressing the debt bias: A comparison between the Belgian and the Italian ACE systems”, Taxation Papers – Working Paper No. 44, European Commission. Additional tables and figures.


Lilas DEMMOU  (✉ lilas.demmou@oecd.org)

Sara CALLIGARIS (✉ sara.calligaris@oecd.org)

Guido FRANCO (✉ guido.franco@oecd.org)

Dennis DLUGOSCH (✉ dennis.dlugosch@oecd.org)

Müge Adalet MCGOWAN (✉ muge.adaletmcgowan@oecd.org)

Sahra SAKHA (✉ sahra.sakha@oecd.org)


← 1. Through a second round effect, the deterioration of assets value could reduce firms’ solvency even more, a channel not investigated in this policy brief.

← 2. Some economically inefficient firms may be categorised mistakenly as efficient and allowed to restructure instead of being liquidated (type-I errors). Conversely, type-II errors occur when some economically efficient firms that encounter temporary financial stress are liquidated, either because existing restructuring procedures are too cumbersome or because the existing procedure mistakenly categorises them as inefficient.

← 3. Along similar lines, a previous policy brief published by the OECD suggests a cascading approach to mitigate liquidity shortfalls and massive insolvency risk (OECD, 2020b).

← 4. On the contrary, calculations do not include support schemes based on loans and loan guarantees as, while critical in addressing liquidity needs, they do not directly contribute to firms’ profitability.

← 5. Consistent with OECD (2020a), to ensure firms’ comparability across countries and sectors, the data are cleaned and prepared as in Gal (2013) and very small firms were excluded to avoid concerns related to the quality of financial statements.

← 6. Reflecting data availability, countries included in the sample are: Belgium, Denmark, Finland, France, Germany, Hungary, Ireland, Italy, Poland, Portugal, Romania, Spain, Sweden and the United Kingdom. At present, Orbis is the largest cross-country firm-level dataset available and accessible for economic and financial research. However, it does not cover the universe of firms, and the extent of the coverage varies considerably across countries. Indeed, in our sample, Italian and Spanish firms add up to half of the observations, while French, Hungarian, Portuguese, Romanian and Swedish firms account, each separately, for 5% to 11% of the observations; all other countries display lower shares. To deal with those limitations, the note purposely avoids in depth cross-country comparisons, as well as the provision of absolute numbers on the aggregate level of the shortfall.

← 7. It is worth noticing that, compared to the cash-flow equation in OECD (2020a), Equation 1 focuses on profits rather than on liquidity, thus excluding non-operating financial commitments (e.g. repayment of the principal of debt).

← 8. Results are qualitatively unchanged with alternative assumptions with respect to the choice of the adjustment variables -- for instance, if we assume that firms deplete their assets to cover losses and increase their liabilities to cover the remaining portion of the decline in profits.

← 9. The full dataset on the confinement shock provided by del Rio-Chanona et al. (2020) can be found here.

← 10. To see why this is the case, consider the following example. Due to confinement measures, a firm is able to produce 50% of its normal time output (e.g. supply shock). If the demand shock, due to changes in consumers’ preferences, implies a 60% reduction in demand for the products of the firm, the firm will produce only what it is able to sell – 40% of its normal time output – and the demand shock is binding. On the contrary, if the reduction in consumers’ demand is expected to be lower (e.g. 20%), the firm will still produce at its maximum capacity during confinement and the supply shock is binding.

← 11. The sectoral implications of the second outbreak characterizing the downside scenario are assumed to be smaller than those of the initial confinement period (e.g. 75% of the size of the initial confinement shock), taking into consideration for example the increased hospital capacity and implementation of better targeted distancing measures.

← 12. As a robustness check, we also computed a more complex structure for the recovery, which is modelled as a combination of the Economic Outlook country-specific projections and sectoral variation obtained from analyst forecasts (IBES data) on the performance of a set of large listed companies. Results are qualitatively unchanged.

← 13. Consistently, between 23% and 29% of previously profitable companies are expected to face losses due to the COVID-19 outbreak.

← 14. Accordingly, 6.2% of previously “safe jobs” are endangered in the upside scenario, while 7.7% in the downside scenario.

← 15. For example, if the assets of the firm are liquidated, the payment accrues first to senior creditors, next to junior creditors and only lastly to equity holders.

← 16. There are several channels through which high leverage can result in debt-overhang: overborrowing during credit booms and the subsequent need for deleveraging to service increased interest payments (Dell’Ariccia et al., 2016), in particular for firms that obtained short-term loans and thus face a high roll-over risk (Acharya et al., 2011); bank-firm relationships, as banks may tend to restrain lending to firms with good growth opportunities in order to keep alive non-viable companies they are exposed to – contributing to reduce the overall level of investment (Caballero et al., 2008; Becker and Ivashina, 2014; Kalemli-Ozcan et al., 2019); the incentive of highly indebted large firms to reduce their activity to meet debt obligations and thus avoid insolvency procedures (Brunnermeier and Krishnamurty, 2020).

← 17. In line with this, there is large evidence that investment collapsed in the aftermath of the global financial crisis and experienced sluggish growth in the following years (OECD, 2020f).

← 18. This includes 90 so-called “fallen angels”, i.e. corporates that lost their investment-grade rating. Furthermore, the cost of corporate refinancing, and therefore corporate investment, are directly exposed to public debt through sovereign credit ratings: since ratings of corporates within a specific country remain at or below the credit rating of the country, any sovereign downgrade could lead to a deterioration of corporate ratings, with adverse consequences on the cost of refinancing, investment and employment (Almeida et al., 2017).

← 19. While this analysis sheds light on the investment-leverage relationship in crisis time, it is worth stressing that the GFC and the COVID-19 shock display very different features (e.g. ranging from the underlying causes to government responses) and thus that results could be interpreted in the light of the current situation only to a limited extent.

← 20. See Table A.3 for detailed results from the regressions.

← 21. In other words, we assume that the increase in the liabilities to total assets ratio is fully driven by an increase in financial debt as the simple accounting model presented in section 1 and data limitations do not allow to calculate the development of the different components of liabilities following the COVID-19 outbreak.

← 22. For a recent discussion on start-ups during the COVID-19 see OECD (2020g). 

← 23. Evaluations of existing systems suggest that ACE systems, if well-designed, reduce leverage at the firm level. In the case of Belgium, the introduction of ACE was associated with a significant decrease in financial leverage (Princenc, 2012, Panier et al., 2013) across larger firms, but no significant changes in the capital structure were found in SMEs (Campenhout and Caneghem, 2013). Empirical results for Italy (Branzole and Caiumo, 2018), Austria (Frühwirth and

Kobialka, 2011) and Turkey (Ozdamar, Tanyeri and Akdeniz, 2020) further support the view of a decrease in leverage.

← 24. For instance, insolvency reforms were more prevalent in countries more severely impacted by the global financial crisis. More recently, in 2020, the UK published a Corporate Insolvency and Governance Bill, which provides new rescue opportunities for struggling companies and to introduce greater flexibility into the insolvency regime, and the Netherlands adopted the Act on Confirmation of Private Plan, which modernises the Dutch insolvency law and introduces fast and flexible restructuring options.

← 25. In particular, the new regime encourages the parties to look for dialogue and out-of-court solutions; provides business with more options for restructuring rather than exit; speeds up court procedures; improves accountability of insolvency administrators; and establishes new supervision rules implying stronger self-regulation.

← 26. For instance, the persistence of such barriers contributed to the increase in the share of zombie firms in the aftermath of the GFC, with negative aggregate productivity effects (Adalet McGowan et al., 2017).

← 27. Although implications are not quite as dire for large corporations, it takes longer for these companies to restructure and costs are substantially higher if the court is congested.

← 28. The availability of a “fresh start” has been found to foster productivity growth via higher incentives for entrepreneurship and experimentation by: i) increasing firm entry; ii) providing failed entrepreneurs with a second chance to apply their experience and lessons learnt to ensure their new businesses grow; and iii) attracting better quality workers – i.e. individuals with higher observed human capital (Adalet McGowan, Andrews and Millot, 2017).


This paper is published under the responsibility of the Secretary-General of the OECD. The opinions expressed and the arguments employed herein do not necessarily reflect the official views of OECD member countries.

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.

© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.