Portugal: Selected Issues

International Monetary Fund. European Dept.

doi:10.5089/9798400215070.002.A001

Author:

International Monetary Fund. European Dept.

International Monetary Fund. European Dept.
Search for other papers by International Monetary Fund. European Dept. in
Current site
Google Scholar
PubMed

Publication Date:: 30 Jun 2022

eISBN:: 9798400215070

ISBN:: 9798400215070

Language:: English

Keywords:: share of firm; firm productivity; lion share; solvency contribution; productivity drag; Productivity; Total factor productivity; Corporate insolvency; COVID-19; Europe

Download PDF (726.2 KB)

1. Prior to the Covid-19 pandemic crisis, a large share of Portuguese firms already exhibited elevated financial risks. Portuguese corporates were on a deleveraging trend after the European sovereign debt crisis in 2012–13 until the 2020 pandemic. On a consolidated basis, NFC sector’s aggregate equity to total liabilities ratio rose from 40 percent in 2011 to 51 percent in 2019, while combined loan and debt security liabilities decreased from 125 percent of GDP in 2012 to 85 percent of GDP in 2019. However, in 2019, some one-third of the firms still did not generate positive net income, a quarter had negative equity (insufficient assets to meet liabilities), and almost one-sixth did not generate enough earnings to cover financing expenses. The risk indicators tended to be weaker in sectors most affected by the pandemic, accommodation and food services in particular. Finally, Portuguese NFCs had relatively weaker financial ratios compared to the euro area (EA) country peers, particularly with regard to share of negative equity firms.

Abstract

How Could The Covid-19 Pandemic Affect Firm Productivity And The Speed Of The Recovery?^¹

A. Firm Demographics and Corporate Financial Health Indicators: Taking Stock of Actual Data

1. Prior to the Covid-19 pandemic crisis, a large share of Portuguese firms already exhibited elevated financial risks. Portuguese corporates were on a deleveraging trend after the European sovereign debt crisis in 2012–13 until the 2020 pandemic. On a consolidated basis, NFC sector’s aggregate equity to total liabilities ratio rose from 40 percent in 2011 to 51 percent in 2019, while combined loan and debt security liabilities decreased from 125 percent of GDP in 2012 to 85 percent of GDP in 2019. However, in 2019, some one-third of the firms still did not generate positive net income, a quarter had negative equity (insufficient assets to meet liabilities), and almost one-sixth did not generate enough earnings to cover financing expenses. The risk indicators tended to be weaker in sectors most affected by the pandemic, accommodation and food services in particular. Finally, Portuguese NFCs had relatively weaker financial ratios compared to the euro area (EA) country peers, particularly with regard to share of negative equity firms.

2. In 2020, the share of firms with negative net income and unable to cover financing expenses out of operating revenues surged, although the solvency picture was more nuanced. According to the 2020 data from the Central Balance Sheet Database of Banco de Portugal (BdP), the pandemic-affected sectors saw a sharp increase in the share of financially weak firms.^² Overall, the share of firms with negative equity rose only marginally, from 25.5 to 26.6 percent. However, firm-level balance sheet data from Orbis, which is used in this analysis, provides a more granular picture at the sectoral level also reflecting the differences in the exit rates of firms between sectors. Specifically, among the firms reporting data for 2020, 6 percent of the firms that had positive equity in 2019 became insolvent in 2020. ^³ This transition into insolvency in the three most affected NACE1 sectors rose to about 14 percent from about 6–8 percent in the previous year. Also, the exit rate rose disproportionately more among the previously solvent firms—the share of insolvent firms that exited increased from 13.7 percent in 2019 to 19.6 percent in 2020, while the share of solvent firms that exited increased from 5.7 percent to 11.9 percent. Consequently, the share of insolvent firms among all firms that exited^⁴ decreased from 44.4 percent in 2019 to 34.9 percent in 2020.

3. On aggregate however, the exit rate was unchanged, while the entry rate dropped sharply. In 2020, the overall exit rate rose from 5.4 percent to 5.7 percent, markedly below the exit rates observed during previous crises. That said, firms’ aggregate birth rate saw a sharp decline to the lowest level since 2009.^⁵ For the affected sectors, while exit rates did not move much, the birth rate fell even more sharply.

4. Overall, pockets of corporate vulnerabilities that emerged in 2020 were masked by improved aggregate NFC balance sheets. Firm level data suggest that the aggregate equity-to-GDP rose in 2020 from about 108 to 116 percent of GDP (equivalent to about 6 billion euros).^⁶ Nonetheless, the aggregate equity gap of insolvent firms deteriorated by about 1 percent of GDP in 2020, mostly accounted for by the most affected sectors. Moreover, among the firms that reported both 2019 and 2020 data, the aggregate equity gap widened by about 2.3 percent of GDP.^⁷ Although the equity gaps narrowed in the information technology and professional services sectors, the widening in the affected sectors (excluding transport) was about 0.8 percent of GDP. As elsewhere in Europe, a handful of large and medium-size companies in the transport sector accounted for the lion share of the equity gaps (3 percent of GDP), dwarfing the deterioration in other parts of the economy. Firms with negative equity of about 2½ percent of GDP in 2019 have not reported thus far 2020 data.

5. A large share of Portuguese firms experienced an increase in leverage and the share of leveraged and insolvent firms increased more than in the EA overall. Portugal was relatively harder hit by the pandemic reflecting its reliance on tourism (GDP fall of 8.4 percent in 2020 vs. 6.5 percent in the EA). In turn, Portuguese NFCs were relatively more affected. The leverage ratio decreased (i.e., improved), in 43 percent of Portuguese firms, compared to 50 percent in the EA. As in the EA, leverage increased in about 1/3 of the firms. However, based on the EU definition of an “undertaking in difficulty” (debt-to-equity ratio exceeding 7.5 applied to all firms or technical insolvency), the share of Portuguese continuing firms with financial difficultly rose from 20 to 28 percent, compared to a rise from 14 to 20 percent in the EA.

6. The share of Portuguese firms unable to cover interest costs by operational income (EBITDA) also jumped from 22 in 2019 to 33 percent in 2020, again somewhat more than in the rest of the EA. Moreover, about one-third of Portuguese continuing firms that had ICR<1 in 2020 also had ICR< 1 in 2019, which suggests that a large share of such firms had a challenging operating income situation even before the pandemic. The share of such firms was also higher in the affected sectors (e.g., 20 percent of the continuing firms in the hotels and food services industry).^⁸ Moreover, as only about a half of nearly 400 thousand Portuguese firms^⁹ represented in Orbis reported interest payments in 2019, among the interest paying firms the share of those with ICR<1 may have jumped to nearly two thirds in 2020.

7. More generally, the NFC sector remains vulnerable to interest rate risks. With nearly 60 percent of NFC loans estimated to be contracted on variable rate with a fixation up to one year^¹⁰, an interest rate rise would amplify corporate cash flow pressures. Firm-level simulations of a 100bps increase in borrowing rates, which is in line with widening market-based benchmark such as EURIBOR swaps since 2020, would increase the share of firms with ICR<1 by 2 percentage points.

B. Assessing Impact of Liquidity and Solvency Support During the Pandemic

8. To estimate the impact of the pandemic on corporate liquidity and solvency we simulate cash flow and equity positions following the methodology of Ebeke and others (2021). Accordingly, we consider a firm as:

Illiquid, if its liquidity position is negative at the end of the period.
Insolvent, if its equity position is negative.

We make several adjustments to the methodology of Ebeke et al (2021). First, we allow firms to partially offset declines in turnover by reducing wage and operating costs, as our focus is on a longer horizon and hence on sustained liquidity shortages rather than the liquidity stress felt during the pandemic. Second, we incorporate the latest policy parameters and turnover outturns on NACE2 level.^¹¹ Third, we distinguish policy uptake by non-distressed and distressed firms, and the liquidity support carry-over into 2021.

9. Credit support measures were key to closing large liquidity shortfalls during the pandemic. Simulations suggest that in the absence of policy support, liquidity strains would have been widespread and especially acute in the most affected sectors. While larger firms were more likely to experience liquidity distress, owing to geared balance sheets, they subsequently are estimated to have benefitted more from policy support. All in all, the share of illiquid firms is smaller post-policies at the end of 2020. Credit measures are estimated to have covered about 6½ percent of GDP out of 8 percent of GDP in crisis-induced liquidity needs (moratoria: 3½ percentage points, credit guarantees: 3percentage points), while job retention programs covered another 2/3 percentage point of GDP. In addition to covering liquidity shortfalls, policies also enhanced NFCs’ cash buffers (credit lines: 3 percent of GDP, other measures: 1 percent of GDP). Lastly, the support provided to liquid firms (on 2020 annual basis) is estimated to be relatively small (3/4 percent of GDP), and primarily attributable to employment support schemes. By making access to finance easier, public support measures may also have helped avoid a more abrupt investment adjustment (Banco de Portugal, 2022).^¹² These findings, nonetheless, indicate that although universal schemes are effective at bridging systemic liquidity shortfalls, they may come with targeting inefficiencies as firms with positive liquidity also benefitted from public support.

10. Nonetheless, policy support closed little of the emerged equity shortfalls, with solvency contributions thus far provided indirectly mainly through job retention and SME turnover loss compensation grants. Our estimates suggest that policies have restored solvency position of around 1½ percent of Portuguese NFCs by the end of 2020, thereby reducing crisis-induced increase in the share of insolvent firms from 6½ to about 5 percent; and by 6 percent of the firms in the accommodation and food sector, thereby reducing the increase in the share of insolvent firms in the sector from 20 to about 14 percentage points. A somewhat lower share of negative equity firms as reported in actual 2020 data (see para. 2) likely reflects unaccounted firm-specific cost-saving factors, and equity injections by the proprietors. Furthermore, the actual data for 2020 corroborates a surge in the share of insolvent firms in the most affected sectors (jumping from about 6–8 percent prior to the pandemic to around 16 percent).

C. Outlook Implications: Distress Indicators and Zombification Risks

11. Corporate distress is expected to have remained stable in 2021. Simulations do not indicate further deterioration in the share of insolvent firms among the firms that were both solvent and liquid prior to the crisis and consequently reported data in 2020. The rise in the share of illiquid firms has been relatively small, which reflects the strong policy support. That said, viability considerations played a limited role in the lifeline programs,^¹³ hence there are risks of an increased share of unviable firms, or those that mask economic and fiscal risks. However, simulations suggest the unwinding of liquidity support may have posed challenges for a small share of firms in 2021 that depleted cash buffers against a still challenging operating environment^¹⁴

12. Related to the above, corporate zombification has risen. The share of zombie companies with (ICR<1 for three consecutive years and firm’s age over 10 year)^¹⁵ is estimated to have risen since the start of the pandemic.^¹⁶ The definition includes three consecutive years of ICR below 1, and hence to be qualified as a zombie in 2021 it requires that a firm had EBIT below its interest bill already before the pandemic. We incorporate actual interest paid by firms in 2020, which among the firms that benefitted from the moratoria may have been temporarily reduced. Recent work by Marques and others (2022) also points to a possible further deterioration of corporate sector health metrics in the near term. Specifically, based on Portuguese firm-level balance sheet simulations, the reduction in the profitability and in the capital ratio of worst-performing firms suggests that heterogeneity in economic recovery may contribute to an increase in corporate insolvency risk, particularly in the most affected sectors. The proportion of firms with negative equity could rise between 2020 and 2023, though to a lesser extent than during the sovereign debt crisis period (2010–2014) and relatively less so on an assets-weighted basis.

D. Risks: Jobs, Economy, Financial System

13. Policy support prevented a severe impact on jobs and the economy. Without policy support, the crisis could have destroyed up to a third of NFC jobs and 20 percent of economic output. Specifically, in the absence of lifelines support (job retention schemes, tax deferrals, credit) the share of NFCs employment by illiquid firms would have jumped from about 15 percent prior to the pandemic to 33 percent. Similarly, on a value-added weighted basis, the share of economic activity of the illiquid firms would have jumped from 10 to 28 percent. As of 2020, lifeline policies are estimated to have significantly helped illiquid firms even compared to pre-Covid-19. Nonetheless, the share of NFC employment (based on pre-Covid-19 employment figures) in firms that ended 2020 with negative equity or were classified as zombies rose slightly. These have also helped ease labor market adjustment. Although in the absence of actual firm-level employment data for 2020, it is yet not possible to infer the extent of employment adjustment, even though aggregate sectoral labor market data suggest reallocation away from the sectors most affected by the pandemic.

14. Although near-term financial stability concerns are thus far contained, risks could grow over time. Support measures, such as moratorium schemes and state-guaranteed credit lines, have limited the materialization of corporate defaults. Moreover, the share of loans to firms with negative equity is estimated to have fallen from 13.3 percent at the end of 2019 to 12.5 percent by the end of 2020, as growth of total loans to NFCs (9 percent) outpaced growth of loans to insolvent firms (3 percent).^¹⁷ At the same time, the share of banking loans to zombie firms has risen from 7 percent pre- Covid-19 to almost 9 percent by the end of 2020 and is likely to have risen to almost 13 percent by the end of 2021 in line with projected increase in the share of zombie firms. Furthermore, both insolvent and zombie firms have significantly increased reliance on short-term debt,^¹⁸ and the share of stage 2 loans that benefitted from the moratoria rose from 17 percent in mid-2020 to 32 percent in December 2021, implying risks of higher eventual insolvencies and NPLs. On a positive note, if the recent lower flow of new loans to such zombie firms is sustained, financial stability risks would remain contained.^¹⁹

E. Dynamism and Allocative Efficiency: Experience from the Past Crisis

15. The Portuguese experience from previous crises suggests that firm dynamism exerts an overall positive “cleansing effect,” though productivity growth differed across firms by size, an effect which may be less prevalent in the current episode. Specifically, while firm dynamism had different overall outcomes during the GFC (2008–12) and EA debt crisis (2013–17) episodes, in both cases, exits contributed to improving total firm productivity. Specifically:

Exit has been an important productivity raising factor throughout both periods (0.33percentage points and 0.15percentage points gain per year), as exiting firms have been significantly less productive. However, during the GFC crisis exiting micro-firms and large companies as a group were more productive than their continuing counterparts, but not for the SMEs. This finding is in line with those of Carreira and Teixeira (2016), who find that financing factors (credit conditions, sales, operating cash flow, leverage), were important determinants of firm exit during the GFC.
While aggregate productivity of micro and SME companies declined (0.45 percentage points per year) during the GFC period of 2008–12, it accelerated to about 1.1 percent per year during the 2013–17 period. This is in line with broad-based economic recovery underpinned by structural reform program following the sovereign debt crisis.
The impact of entry created a drag on aggregate productivity in both periods. This appears to have been primarily due to compositional factors. Although entrants were more productive within their respective groups, new micro firms tend to less productive than continuing SMEs hence dragging down aggregate contribution of entry.
In contrast with 2008–12 period, continuing firms contributed almost 1 percentage point per year to TFP growth during 2013–17, of which about two thirds were on account of within firm productivity growth.
Lastly, the two crises also differed in terms of productivity change among large firms. TFP of large firms was nearly flat during 2008–13, with positive contribution from within firm gains (0.10 percentage points per year) offset by allocative losses (0.13 percentage points per year). In turn, during 2013–17, TFP of large firms grew by almost 3 percentage points per year, with gains on all components, and dominance of allocative efficiency and within firm productivity.
As regards the current episode, using Portuguese firm-level data survey of firms matched with administrative data, Kozeniauskas and others (2022) find that there has been no rise in exit among lower-productivity firms. This is in line with theory that support policies offset the cleansing effect of recessions. They also find that high-productivity firms have been less likely to take up government support.

16. Qualitatively, the results are comparable for other European countries and the analysis of Portuguese firms’ dynamics during the GFC. For example, Patnam (2020) finds a comparable drop in TFP for French SMEs during 2008–12 driven by declines of within firm productivity, which is partly offset by exit and to a much smaller extent by gains from entry and reallocation. In the case of Portugal, Carreira and Texeira (2016) find a negative within-firm effect and a positive effect exerted by resource reallocation and entering firms in Portugal during the 2008–12 crisis. Our results suggest a similar exit dynamic during the GFC, with cleansing exit dynamics dampened by the shutdown of relatively more productive firms among the groups of micro and large enterprises. These results complement the recent Banco de Portugal (2021b) study, focused on within sector employment-weighted productivity and labor reallocation channel,^²⁰ by allowing across-sector and multi-factor reallocation dynamics.

17. The productivity drag from continuing firms with high debt exerts a sizable additional drag on aggregate productivity.^²¹^, ^²² The primary channel of zombie’s drag on productivity growth comes from allocative efficiency losses.

In the 2008–12 period, the total incumbent firms’ productivity loss of 0.49 percentage points can be decomposed into loss of productivity of 0.28 from non-zombie firms, and 0.21 percentage points among the zombies. The latter implies a material drag on aggregate productivity, given that zombie enterprises represent only about 6 percent of total. Moreover, non-zombie productivity change has been mostly on account of resource allocation losses. Furthermore, the positive non-zombie within firm productivity change is attributed to micro firms even though within firm change is negative for non-zombie SMEs. This may suggest that in addition to overall allocative efficiency challenges during the GFC, larger SMEs faced further internal adjustment difficulties, likely in view of the prominent past labor market and insolvency regime rigidities. In contrast, the large non-zombie enterprises were able to contribute positively both via allocation and internal productivity.
In the 2013–17 period, the total incumbent firms’ productivity gain of 0.98 percentage points can be decomposed into gain of productivity of 1.31 percentage points from non-zombie firms, and 0.33 percentage points loss among the zombies. While the negative contribution of within firm productivity change among zombies had a negligible impact overall, they posed a material drag of 0.31 percentage points per year through allocative efficiency as they kept resources locked from flowing towards more productive non-zombie firms. Critically, the productivity growth differential between zombies and non-zombies had been substantial during this period.

18. The results mirror evidence on debt overhang issues in the European context and pose critical implications for corporate solvency policy. For instance, Duval and others (2020) show that firms with weaker balance sheets experienced a highly persistent decline in post-crisis productivity growth accounting for about one-third of within-firm productivity slowdown. Moreover, firms with higher leverage reduce investment more (Kalemli-Oczan et al. 2018, Demmou et. al, 2020). Overall, comparing the two crisis episodes reveals that zombie firm chip away at aggregate productivity irrespective of the performance of non-zombie incumbents and congest reallocation. Moreover, the productivity gains during 2013–17 period which saw a TFP recovery suggest that most of the gains come from the continuing non-zombie firms as well cleansing Schumpeterian effects of entry and exit.

19. The rise in the share of zombie firms could chip away at aggregate TFP growth 0.2–0.4 percentage points per year over the medium-term. As indicated in the previous section, the share of zombie is estimated to have risen by nearly 3 percentage points. Based on estimates of the productivity drag from zombies during the past two crises, the allocative efficiency and productivity losses prevalent among zombie firms could chip away at aggregate TFP growth 0.2–0.4 percentage points per year over the medium-term.

F. Policy Conclusions

20. The pandemic eroded equity positions of Portuguese firms with varying impacts across firms’ size and sectors. Although the economy-wide share of negative equity firms has risen only marginally (2 percentage points), the share of pre-Covid-19 solvent firms that turned insolvent by the end of 2020 surged in the affected sectors. Moreover, shares of firms with negative net income and unable to cover financing expenses out of operating revenues also increased.

21. The authorities’ swift policy response went a long way in addressing immediate liquidity shortages. Liquidity support, primarily via moratoria and credit lines, provided firms with sufficient liquidity buffer. Support to solvency has, however, been small thus far. Notwithstanding targeting of support thus far, SMEs and micro enterprises remain vulnerable, while over a third of companies in the affected sectors, such as food and hospitality and arts and recreation, are technically insolvent.

22. The pandemic, however, has left a large share of Portuguese corporates with a debt overhang and at risk of insolvency. Our estimates suggest that the share of zombie firms has risen from about 1 percent prior to the pandemic to 4 percent. The solvency picture is more nuanced as pockets of corporate vulnerabilities that emerged in 2020 have been masked by improved aggregate NFC balance sheets. Nonetheless, among the firms that reported both 2019 and 2020 data, the aggregate equity gap widened by about 2.3 percent of GDP. Firms that were insolvent or were classified as zombies by end of 2020 saw doubling of the ratio of their short-term to total loan liabilities, from about a quarter pre-Covid-19 to almost a half. For the affected sectors, this risk could be higher, especially if the economic recovery falters or the most affected sectors remain under pressure. Additional vulnerabilities due to the war in Ukraine, cost-push pressures, supply-chain disruptions and higher interest rate could elevate insolvency risks.

23. Although targeted support, via debt-equity swaps and capital injections, to the transport-related SOEs have helped with large solvency gaps, enhanced governance and viability considerations will be key to reducing fiscal risks. State capital injections into main transport SOE (the Lisbon and Porto subways or the national rail network) amounted to 1 percent of GDP in both 2020 and 2021 with support to the national airlines, including under the restructuring plan, is expected to be close to 3 percent of GDP. The pandemic has compounded pre-existing risks to the financial sustainability of these companies, and it is critical that efforts to revitalize balance sheet are complemented by measures to address governance challenges.

24. The Banco Portuguese de Fomento managed recapitalization scheme—Strategic Recapitalization and Consolidar Programs—have many promising features but may need (size and instrument) augmentation and enhanced incentive structures. The Consolidar program offers new venues for tapping expertise and capital of private institutions (venture capital and equity funds in the case of Consolidar), although the envelope may require augmentation in view of relatively large equity shortfall among affected SMEs and uncertainties surrounding the tourismrecovery. Program flexibility is critical, and a quantitative evaluation desirable as new information about take-up rates, implementation challenges, the strength of economic recovery, and the ability of the program to stabilize firms becomes available.

25. Broad-based economic recovery underpinned by structural reforms would help spur firm dynamism, productivity growth, and strengthen financial health metrics. Continued strong growth would bolster the operational environment across the spectrum and help reduce balance sheets vulnerabilities. Nonetheless, past crisis recoveries reveal that zombie firms chip away at aggregate productivity, irrespective of the performance of non-zombie incumbents, thereby congesting reallocation. Strong restructuring and insolvency regimes would facilitate effective reorganization and exit of business and optimizing resource reallocation without overwhelming the financial system.

Annex I. Technical Notes

Balance Sheet Simulations—Technical Notes

Liquidity position is simulated according to:

\begin{matrix} C a s h_{t} = C a s h F l o w_{t} + L i q u i d A s s e t s_{t} - M a t u r i n g L i a b i l i t i e s_{t} - D i v i d e n t_{t} + N e w C r e d i t_{t} - I n t e r e s t_{t} N e w C r e d i t_{t} & (2) \end{matrix}

\begin{matrix} C a s h F l o w_{t} = (1 - β_{j}) (S a l e s_{t} - M a t e r i a l C o s t_{t}) - (1 - γ_{j}) W a g e s_{t} - O t h e r O p e r a t i n g C o s t s_{t} + (F i n Re v e n u e_{t} - F i n E x p e n s e_{t}) - I n t e r e s t E x p e n c e_{t} - T a x e s_{t} & (3) \end{matrix}

Where β_j is industry-specific turnover shock, and γ_j is the extent of industry-specific wage bill adjustment. The turnover shock reduces operational cash-flows of firms in sector) through a decline in sales. Firms are also able to adjust material costs in the same proportion as the change in turnover, but the extent of employment adjustment varies according to γ_j. Compared to the methodology of Ebeke et al (2021), we effectively allow firms to partially offset declines in turnover via reducing wage costs. Underlying to this difference is focus on a longer horizon and hence sustained liquidity shortages rather than the extent of liquidity stress felt during the acute stage of the first several months of the pandemic. Firms need to meet obligations on fixed costs, financial payments, and corporate taxes. On the revenue side, they receive financial revenue from financial investments. Hence, we assume that firms continue to generate cash flows via sales as opposed to receivables, while they also face cash outflow through purchase of supplies rather than trade payables. Lastly, we impose several stages of the cash flow dynamics (Figure), in order to distinguish firm’s use of policy support, such as credit moratoria, consequent use of credit lines, as well as impact of tax payment liabilities or tax loss carry-over in the outer years.

Equity position is then calculated according to:

\begin{matrix} E q u i t y_{t} = E q u i t y_{t - 1} + C a s h F l o w_{t} - A m o r t i z a t i o n_{t} - I n t e r e s t_{t} N e w C r e d i t_{t} + S o l v e n c y B o o s t_{t} & (4) \end{matrix}

Where SolvencyBoost_t denotes solvency support element associated with non-refundable lifeline policy support, which de facto is limited to (i) job retention and wage subsidy schemes including corresponding social security contribution exemptions, and (ii) SME grant provided under APOIAR.PT. All other policies are considered to only affect liquidity (Table 1).

^{Table 1.}

Key Policy Support Measures

Note: Policy impact: L=Liquidity support, S = solvency support * estimate of debt service affected by moratoria based on eligibility and liquidity shortfall (non-precautionary). No corresponding figures of debt service are officially reported. ** for 2021, estimate of uptake is lower than the envelope.

The 2021 forecast incorporates NACE1-level recovery to normal time responses from Phase 3 of Covid-19 WB Enterprise Surveys (February 2021), adjusted for turnover outrun through mid-2021. The average recovery time across sectors of economic activity equals 14 months and implies economy-wide recovery to pre-pandemic activity level by the first quarter of 2022, which has been generally in line with the Survey’s expectations. Further, wage bill adjustment equals NACE2 employment change. The analysis uses firm-level data on balance sheets and income statement from the Orbis database. The simulation covers about 250 thousand firms operating in Portugal during 2018, taking the latest financial statements available for each firm. Although this represents about a third of registered companies, with micro firms slightly underrepresented, it represents almost 80 percent of total operating revenue of corporates in Portugal. To ensure comparability and representativeness of the results, particularly assessment of the economy-wide liquidity and equity shortfall, we upscale the sample using OECD Structural Demographics and Business Statistics by aggregate turnover at the NACE2 and company size.

Productivity Decomposition—Technical Notes

Changes to aggregate productivity can be decomposed into the firm-specific factors as well as factors related to dynamism of firm entry and exit. Using data on SMEs from ORBIS^¹ for estimation, the change of aggregate productivity, measured using the Levinsohn-Petrin (2003) method, is decomposed following Melitz and Polanec (2015):

\begin{matrix} Δ P_{t} = Δ {\bar{P}}_{C t} + Δ cov (θ_{i t}, p_{i t}) + θ_{E t} (P_{E t} - P_{C t}) + θ_{X t} (P_{C (t - τ)} - P_{X (t - τ)}) & (1) \end{matrix}

Where P_t represents the aggregate productivity level in year t, and C, E, and X denote the groups of continuing, entering, and exiting firms; θ_it and p_it is the firm’s value-added market share and productivity level, respectively. θ_Gt is the share of group G; and P_Gt and ${\bar{P}}_{G t}$ are correspondingly value-added weighted and unweighted average productivity for each of the groups G = (C,E,X). Correspondingly, the first term captures the contribution of within-firm productivity changes of continuing firms. The second term reflects inter-firm resource reallocation towards more productive continuing firms. The last two terms capture the aggregate productivity contribution of entering and exiting firms, respectively.

References

Banco de Portugal, 2021a, Financial Stability Report, December 2021.
- Search Google Scholar
- Export Citation
Banco de Portugal 2021b, Dynamics of productivity per worker in Portuguese firms over the 2014–19 period, Economic Bulletin December 2021.
- Search Google Scholar
- Export Citation
Banco de Portugal 2022, The Evolution of firm’s liquidity and leverage in 2020, Economic Bulletin May 2022.
- Search Google Scholar
- Export Citation
Ebeke C., Jovanovic N., Valderrama V., and Zhou J., 2021, Corporate Liquidity and Solvency in Europe during Covid-19: The Role of Policies, IMF Working Paper 21/56.
- Search Google Scholar
- Export Citation
Carreira, C. and Teixeira, P., 2016. Entry and exit in severe recessions: lessons from the 2008–2013 Portuguese economic crisis. Small Business Economics, 46(4), pp.591–617.
- Search Google Scholar
- Export Citation
Carreira, C., Teixeira, P., and Nieto-Carrillo E., 2021. Recovery and exit of zombie firms in Portugal. Small Business Economics, https://doi.org/10.1007/s11187–021-00483–8.
- Search Google Scholar
- Export Citation
Duval, R., Hong, G.H. and Timmer, Y., 2020. Financial frictions and the great productivity slowdown. The Review of Financial Studies, 33(2), pp.475–503.
- Search Google Scholar
- Export Citation
Gopinath, G., S. Kalemli-Ozcan, L. Karabarbounis, and C. Villegas-Sanchez 2017. Capital Allocation and Productivity in South Europe.” Quarterly Journal of Economics 132 (4): 1915–1967.
- Search Google Scholar
- Export Citation
Kalemli-Ozcan, S., B. Sorensen, C. Villegas-Sanchez, V. Volosovych, and S. Yesiltas, 2015. How to Construct Nationally Representative Firm Level Data from the ORBIS Global Database. NBER Working Paper No. 21558.
- Search Google Scholar
- Export Citation
Kalemli-Ozcan, S., Laeven, L. and Moreno, D., 2018. Debt overhang, rollover risk, and corporate investment: Evidence from the european crisis (No. w24555). National Bureau of Economic Research.
- Search Google Scholar
- Export Citation
Koseniauskas, N., P. Moreira, 2022, On the cleansing effect of recessions and government policy: Evidence from Covid-19, European Economic Review 144
- Search Google Scholar
- Export Citation
Levinsohn, J. and Petrin, A., 2003. Estimating production functions using inputs to control for unobservables. The Review of economic studies, 70(2), pp.317–341.
- Search Google Scholar
- Export Citation
Marques, C., F. Augusto, R. Martinho, 2022. Modelling the financial situation of Portuguese firms using micro-data: a simulation for the Covid-19 pandemic, Banco de Portugal Occasional Paper 3/2022.
- Search Google Scholar
- Export Citation
Melitz, M.J. and Polanec , S., 2015. Dynamic Olley-Pakes productivity decomposition with entry and exit. The Rand Journal of economics, 46(2), pp.362–375.
- Search Google Scholar
- Export Citation
McGowan, M. A., and rews, D., and Millot V, 2018, “The walking dead? Zombie firms and productivity performance in OECD countries”, Economic Policy, 33(96), pp. 685–736
- Search Google Scholar
- Export Citation

Prepared by Lakshita Jain (University of North Carolina at Chapel Hill) and Volodymyr Tulin (EUR).

Specifically in the accommodation and food service activities, the share of loss-making (i.e., negative net income) firms jumped from an already high of 45½ percent to 67½ percent, the share of firms unable to cover financing expense with operating revenues rose from 19 to 34 percent, and the share of firms with negative equity rose from 40 to 43 percent.

Throughout this paper insolvent signifies negative equity position (shareholder funds) on firm’s balance sheet.

Although lack of 2020 Orbis data reporting does not strictly indicate an exit, aggregate share of firms without 2020 data at 5.9 percent matches well firm closure rates of 5.7 percent in the BdP’s Central Balance Sheet Database.

Eurostat data for 2020 indicates a comparable large drop of 24 percent in new business registration and a small increase of about 3 percent in bankruptcy filings compared to 2019. Moreover, bankruptcy filings have been on a downward trend since the 2020:Q1, dropping about 1/4^th as of 2022:Q1 relative to the 2019 average. In 2022:Q1, new business registration increased by 22 percent relative to 2019.

Simulations for 2021, discussed below, indicate further aggregate balance sheet improvement and further widening of corporate negative equity under the assumptions of no firm exit and no new external equity support which likely overstate the extent of solvency gap deterioration.

The deterioration in aggregate solvency gap is lower primarily due to survivorship bias with non-reporting of 2020 data by firms with solvency gaps in 2019.

As reported in the December 2021 Financial Stability Report (Banco de Portugal, 2021a), the financial vulnerability indicator using the proportion of operating income allocated to interest payments in each firm, debt of vulnerable firms in the most affected sectors doubled between 2019 and 2020, while firms that were financial vulnerable in 2020 increased their debt by 30 percent.

Orbis sample covers about 99 percent of NFC turnover (2019, relative to BdP’s Central Balance Sheet Database), and 72 percent of employment (2018, relative to OECD Annual Labor Force Statistics),

Based on ECB Risk Assessment Indicator Database.

The turnover assumptions entail recovery to normal time as per responses from Phase 3 of Covid-19 World Bank Enterprise Surveys (February 2021) at NACE1-level adjusted for turnover outrun through mid- 2021. Specifically, the sectoral average recovery time equals 14 months which implies economy-wide recovery to pre-pandemic level by the first quarter of 2022, though with variation

Drop in investment also played a key role in cash flow adjustment. Moreover, the fall in investment was sharper in firms where a greater observed decrease in cash flow from operations (Banco de Portugal, 2022).

While credit guarantee programs entailed risk mitigating access qualification requirements on the financial situations of a firm, such as positive net position, these requirements were backward looking.

The 2021 balance sheet simulations may likely overstate the corporate distress indicators due to unaccounted cost savings factors or equity injections.

ICR is based on EBIT, given its more common use in literature as measure of operating income. The choice of the profit measure (EBIT vs. EBITDA) affects the calculated ICR and consequently the incidence of zombie firms (EBIT deducts the non-cash expenses related to depreciation and amortization from net profit, whereas EBITDA does not). Although depreciation and amortization are not actual cash outflows, they reduce the value of a company’s capital and/or financial assets and thus of its total assets. EBITDA might be suitable for international comparisons given cross-country differences in depreciation or amortization practices, treatment of goodwill or taxation that may distort bottom line comparability. However, EBIT is more suitable for country-specific analysis due to capturing the different effects of depreciation and amortization on companies (or industries) with different capital intensity use.

In line with Banco de Portugal (2021a), the results indicate decline in zombification rate since the sovereign debt crisis. For example, also based on McGowan et al. (2018) methodology, BdP reports a share of zombie firms of 6.9 percent in 2019, a decline of 4 percentage points from the sovereign debt crisis peak, and a somewhat lower incidence based on asset-weighted metric (decline from 7.7 percent to 2.9 percent in 2019). Lower zombification reported in our analysis should reflect greater coverage of micro firms in BdP’s Central Balance Sheet Database.

Although credit lines which accounted for about 12 percentage point increase in bank loans did not require solvency, credit growth for such companies appears to have been modest.

The two groups overlap. Specifically, about half of zombie firms are also insolvent, but only a small fraction of insolvent firms (4 percent in 2019 and 6 percent in 2020) are classified as zombies.

Banco de Portugal (2021a) reports that a relatively low share of new loans (with and without state guarantees) were granted to zombie firms during the pandemic and a greater share of financially weaker firms among those that opted for moratoria, given softer access requirements.

BdP’s result indicate that reduced employment share of firms with higher productivity and the less favorable evolution of the productivity of firms with a higher employment share have had a highly negative impact on this reallocation channel within industries.

The results qualitatively are robust to alternative zombie classification, such as a higher ICR threshold of 1.5 and potential industry-specific heterogeneity via an additional zombie requirement of having an above median leverage. The preference for 10-year age is more conservative to 5-year on firm maturity grounds. For example, the 5-year age threshold is the age limit defined by the OECD for young high-growth firms, while most studies point out that firms achieve the mature state somewhere between the sixth and tenth year of existence (Carreira & Teixeira, 2011). Carreira and Teixeira (2021) also suggest that there are no major changes to Portuguese firm zombie classification using 5- or 10-year age limit.

Zombies are less productive and employ more people. Among the SMEs, the average employment was 29 people compared to 22 among the non-zombie firms. Although aggregate SME productivity differential stood at close to 10 percent, zombies tend to be some 20 to 40 percent less productive within their respective industries.

ORBIS data were cleaned following steps that are based on Kalemli-Ozcan et al (2015) and Gopinath et al (2017). Micro firms and SMEs i.e., employing less than 250 persons or annual turnover below 50 mln euro, comprise 99.9 percent of firms in the non-financial sector, and generate 55.8 percent of value added and 64.1 percent of employment. We exclude the following NACE1: Education, Financial and insurance activities, Administrative and support service, Public administration and defense, and Human health and social work activities.

Other Publishers

Save
Cite
Email this content

Share Link

Copy this link, or click below to email it to a friend
Email this content
or copy the link directly:

https://www.elibrary.imf.org/view/journals/002/2022/204/article-A001-en.xml
The link was not copied. Your current browser may not support copying via this button.

Link copied successfully

Collapse
Expand

Portugal: Selected Issues

Author:

International Monetary Fund. European Dept.

Volume/Issue:: Volume 2022: Issue 204

Publisher:: International Monetary Fund

ISBN:: 9798400215070

ISSN:: 1934-7685

Pages:: 44

DOI:: https://doi.org/10.5089/9798400215070.002

Keywords
How Could The Covid-19 Pandemic Affect Firm Productivity And The Speed Of The Recovery?1
Annex I. Technical Notes
- Balance Sheet Simulations—Technical Notes
- Productivity Decomposition—Technical Notes
References

View raw image

Pre-pandemic Corporate Financial Risk Indicators, 2019

(In percent of firms)

View raw image

Vulnerable Firms, 2019

(percent of all firms)

View raw image

Insolvent Firms: Sectoral Share

(percent all firms, i.e. not just with data in both 2019 and 2020)

View raw image

Corporate Financial Risk Indicators, 2016–20

(In percent of firms)

View raw image

Share of 2018 Solvent Firms That Turned Insolvent in 2019

(percent)

View raw image

Share of 2019 Solvent Firms That Turned Insolvent in 2020

(percent)

View raw image

NFC’s Birth and Death Rates, 2007–2020

(In percent of firms)

View raw image

Aggregate Equity of Nonfinancial Firms

(In percent of GDP)

View raw image

Equity Gaps of Reporting Firms

(bn euro, only firms with data in both 2019 and 2020 are included)

View raw image

Portugal: Share of Firms by Financial Situation (2020–2019)

(percent of reporting continuing firms, financial diffculty is if debt-to-equity is >7.5)

View raw image

Euro Area: Share of Firms by Financial Situation (2020–2019)

(percent of reporting continuing firms, financial diffculty is if debt-to-equity is >7.5)

View raw image

Portugal: ICR<1 Firms

(percent, firms reporting in both 2019 and 2020)

View raw image

Share of Firms with ICR<1

(percent, only firms with data in both 2019 and 2020 are included)

View raw image

Share of Illiquid Firms

(In percent of firms)

View raw image

Share of Insolvent Firms

(In percent of firms)

View raw image

Corporate Distress Indicators: 2021

(In percent of firms)

View raw image

Zombie Firms

(In percent of total firms)

View raw image

Employment by Insolvent and Zombie Firms

(In percent of NFC employment)

View raw image

Employment Adjustment

(Q4/Q4 percent change, for 2021: 2021Q3 over 2020Q4)

View raw image

Vulnerable Firms: Net Liquidity Position

(Ratio of Cash and equivalents to Short-Term Debt, eop)

View raw image

Loans to Insolvent and Zombie Firms

(In percent of total bank loans to NFCs)

View raw image

Vulnerable Firms: Share of Short-Term Credit

(In percent of total bank loans to insolvent or zombie firms)

View raw image

TFP Growth Decomposition: 2008–12, SMEs and Micro Firms

(In percentage points per year)

View raw image

TFP Growth Decomposition: 2013–18, SMEs and Micro Firms

(In percentage points per year)

View raw image

TFP Growth Decomposition: 2008–12 Role of Zombies: SMEs and Micro Firms

(In percentage points per year)

View raw image

TFP Growth Decomposition: 2013–17 Role of Zombies: SMEs and Micro Firms

(In percentage points per year)

View raw image

	2020		2021
	size, % GDP	impact	size, % GDP	impact
Job retention schemes	0.7	L, 5	0.5	L, S
(STW. progressive recovery, wage subsidy)
Social security contribution exemption	0.25	L, 5	0 1	L, S
SME grants (turnover loss compensation)	...		0.5	L, S
*Policies with solvency support:*	0.95		1.1
Corporate tax deferral	0.3	L	...	L
Bank loan sen/ice moratoria, NFCs *	3.5	L	2.5–4.0	L
Loan guarantee schemes ”	4.5	L	0.5	L
*liquidity support policy total:*	8.3		3–4.5

	2020		2021
	size, % GDP	impact	size, % GDP	impact
Job retention schemes	0.7	L, 5	0.5	L, S
(STW. progressive recovery, wage subsidy)
Social security contribution exemption	0.25	L, 5	0 1	L, S
SME grants (turnover loss compensation)	...		0.5	L, S
*Policies with solvency support:*	0.95		1.1
Corporate tax deferral	0.3	L	...	L
Bank loan sen/ice moratoria, NFCs *	3.5	L	2.5–4.0	L
Loan guarantee schemes ”	4.5	L	0.5	L
*liquidity support policy total:*	8.3		3–4.5

	2020		2021
	size, % GDP	impact	size, % GDP	impact
Job retention schemes	0.7	L, 5	0.5	L, S
(STW. progressive recovery, wage subsidy)
Social security contribution exemption	0.25	L, 5	0 1	L, S
SME grants (turnover loss compensation)	...		0.5	L, S
*Policies with solvency support:*	0.95		1.1
Corporate tax deferral	0.3	L	...	L
Bank loan sen/ice moratoria, NFCs *	3.5	L	2.5–4.0	L
Loan guarantee schemes ”	4.5	L	0.5	L
*liquidity support policy total:*	8.3		3–4.5

Topics

Countries

Series

About

Resources

Abstract

How Could The Covid-19 Pandemic Affect Firm Productivity And The Speed Of The Recovery?1

A. Firm Demographics and Corporate Financial Health Indicators: Taking Stock of Actual Data

B. Assessing Impact of Liquidity and Solvency Support During the Pandemic

C. Outlook Implications: Distress Indicators and Zombification Risks

D. Risks: Jobs, Economy, Financial System

E. Dynamism and Allocative Efficiency: Experience from the Past Crisis

F. Policy Conclusions

Annex I. Technical Notes

Balance Sheet Simulations—Technical Notes

Productivity Decomposition—Technical Notes

References

Same Series

Other IMF Content

Other Publishers

Asian Development Bank

Inter-American Development Bank

International Labour Organization

The World Bank

Share Link

Table of Contents

Headings

Figures

Metrics

Foresee

How Could The Covid-19 Pandemic Affect Firm Productivity And The Speed Of The Recovery?^¹