Abstract
Italy: Selected Issues
The Italian and Spanish Corporate Sectors in the Aftermath of the Crisis1
A. Introduction
1. The nonfinancial corporate sectors (NFC) in Italy and Spain were hit hard by the global financial crisis and the economic downturn that followed. In Italy, firm profits fell sharply, and debt levels, although not particularly high at the onset of the crisis, have risen relative to income (Figure 1). In contrast, the Spanish corporate sector borrowed heavily during the boom years with the NFC debt to GDP ratio peaking at nearly 190 percent in 2007. Since then, Spanish NFCs have deleveraged quite rapidly, but debt remains high, at nearly 150 percent of GDP. The significant debt overhang in both countries poses risks to the recovery despite mitigating policy actions, and has led to a significant deterioration in the asset quality of the banking systems. At 17 percent, the nonperforming loan (NPL) ratio in Italy has reached systemic levels, impeding the supply of credit, and potentially holding back private investment. In Spain, the stock of NPLs has started to decline, reflecting concerted efforts to cleanse the banking sector balance sheets. But at 12.5 percent, the NPL ratio remains high. Debt overhang and impaired balanced sheets could act as a drag on firm investment. Indeed, real private investment has fallen to levels not seen in 15 years, dramatically slowing the pace of recovery, and casting a shadow on these economies’ long run potential output. This is of particular concern in Italy, where investment was weak even before the crisis.
Corporate Debt
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Eurostat, National authorties, Haver Analytics. IMF staff calculation.1/ Data on credit growth for Spain is based on BdE’s Financial Variables statistics (table 8.5, series 6.)Corporate Debt
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Eurostat, National authorties, Haver Analytics. IMF staff calculation.1/ Data on credit growth for Spain is based on BdE’s Financial Variables statistics (table 8.5, series 6.)Corporate Debt
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Eurostat, National authorties, Haver Analytics. IMF staff calculation.1/ Data on credit growth for Spain is based on BdE’s Financial Variables statistics (table 8.5, series 6.)2. This paper evaluates the health of the Italian and the Spanish corporate sectors using firm level data, the implication for the strength of banks’ balance sheets, and impact on firm investment. In particular, we ask the following questions: (i) What is the state of the Italian and the Spanish nonfinancial corporate sectors, in terms of profitability, liquidity, leverage, cost of borrowing, and debt repayment capacity? What have the main trends been over the past 8 years, across firms of different size, and economic sectors? (ii) How vulnerable are the corporate sectors of Italy and Spain to an adverse shock to real activity, or a rise in funding costs? What would be the likely impact of a rebound in activity on firms in poor financial health? And (iii) To what extent have impaired balance sheets of firms and banks held back corporate investment in Italy and Spain?
3. The rest of the paper is organized as follows: Section B describes the firm level data and methodology used in the analysis. Section C takes stock of the health of firms in Italy and Spain, and examines their vulnerability to various shocks. Section D presents econometric evidence linking firm’s balance sheets and credit conditions to corporate investment decisions, while Section E offers policy considerations and concludes.
B. Data and Methodology
4. The paper uses data for roughly one million Italian companies and about 700,000 Spanish ones for the 2006–13 period.2 The Orbis database, compiled by Bureau van Dijk, includes all companies required to submit accounts with the Chambers of Commerce in Italy and the Registered Commerce of the Province in Spain. It thus includes a very significant portion of the micro, small, and medium enterprises, which constitute the bulk of economic activity in Italy and Spain (Figure 2).3 These types of firms are rarely represented in other commonly-used firm level datasets. The coverage in the Orbis database is high: the firms included in the database account for roughly 70 percent of the gross value added, and 75 percent of the total wage bill of Italy’s NFC. In Spain, the coverage of the Orbis sample is also high, accounting for half to about 2/3 of the gross value added and cost of employees.4
SME Demographics, 2012
(Number of firms per 1,000 population)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Source: Eurostat.SME Demographics, 2012
(Number of firms per 1,000 population)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Source: Eurostat.SME Demographics, 2012
(Number of firms per 1,000 population)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Source: Eurostat.5. We gauge corporate health and vulnerability based on several commonly used indicators of profitability, leverage, and ability to service debt. Using the detailed, harmonized balance sheets, and profit and loss statements, we construct firms’ return on assets and profit margins, their debt-to-assets and debt-to-income ratio, and the interest coverage ratio (ICR). Having described how the financial state of the corporate sector has evolved since 2006, we classify firms (and the debt that they hold) as vulnerable if, for example, their current revenues are insufficient to fund interest expenses (ICR<1). An ICR below one in a particular year does not indicate that insolvency is imminent. Firms may have assets that can be easily liquidated, unused credit lines, or other sources of funding on which to draw. But research has found that such indicators could be particularly good harbingers of vulnerabilities.5 Stress test of the 2013 corporate balance sheets, with shocks to borrowing costs, and profits (on the order of the changes already observed since the beginning of the global financial crisis) shed light on the likely effect of worsening (or improving) economic conditions on firms’ financial health.
6. We then use the findings on corporate sector vulnerability as indicators of the prospective debt at risk in the banking system in alternative scenarios. The ICR can illustrate in a simplified way the links between firms’ financial performance and the financial system’s asset quality. If macroeconomic conditions deteriorated and firms with low ICR could not find additional sources of funds, they would likely delay interest payments. If the delay persisted, loans to those companies would eventually be classified as nonperforming assets in banks’ balance sheets. By aggregating the financial liabilities of firms with ICRs below a particular threshold, and comparing this number to the total financial liabilities of the firms in our sample, we get a rough estimate of how large corporate debt-at-risk could be in alternative scenarios.6 In this thought exercise, the corporate financial link is particularly important in Italy due to the structural characteristics of the economy. Small businesses, which form the backbone of the Italian economy, are particularly reliant on bank credit due to limited access to capital market funding. Indeed, Italy has one of the highest shares of domestic bank loans in total financial debt of the nonfinancial corporate sector (Figure 3). Italian banks also have a strong focus on traditional deposit-taking and lending activities, and more than 65 percent of total bank credit goes to the corporate sector.
Domestic Bank Loans in the Non-Financial Corporate Sector
(Share of Total Financial Debt, end 2013)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: European Commission, Italy Country Report, February 2015.Domestic Bank Loans in the Non-Financial Corporate Sector
(Share of Total Financial Debt, end 2013)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: European Commission, Italy Country Report, February 2015.Domestic Bank Loans in the Non-Financial Corporate Sector
(Share of Total Financial Debt, end 2013)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: European Commission, Italy Country Report, February 2015.7. Finally, to examine the link between impaired firm balance sheets, credit conditions, and corporate investment, we estimate a standard investment model. We exploit the variation over time and across firms in balance sheet strength, debt overhang, and profitability to estimate their effect on firm investment, and how this effect might have changed in the postcrisis period.7 In the case of Italy, we also look for evidence of financial frictions operating through the banking sector, by examining whether firms in sectors more dependent on external finance and in regions where lending standards tightened the most or credit contracted more severely experienced a sharper drop in investment.
C. The Health of the Corporate Sector in Italy and Spain
8. The crisis has left its mark on the health of the corporate sector in Italy and Spain. Evidence of weakening performance can be seen across a range of measures of profitability and liquidity. The deterioration has occurred across the board, though the negative effects of the crisis are particularly pronounced for smaller businesses, and firms in select sectors, such as construction. Below we describe the time trends of key financial ratios, based on a sample of firms that have been active throughout 2006–13. Focusing on surviving firms allows us to abstract from the compositional changes that might have occurred during this time period due to firm entry and exit, which may not be very well captured in the Orbis database. This comes at the cost of potentially introducing “survivor” bias in some of the results.8 The reported summary statistics are based on data excluding the top and bottom 5th percentile so as to avoid distortions from extreme outliers. Throughout the paper, we use the European Commission classification of firms by size according to their assets for Italy and according to the number of employees for Spain.9 Micro firms are those with total assets of less than 2 billion euros or less than 9 employees. Small and medium (SME) firms have total assets between 2 and 43 billion euros or between 10–249 employees, while large firms are those with assets exceeding 43 billion euros or with more than 250 employees.
Profitability
9. The profitability of Italian and Spanish firms fell dramatically over 2006–13. For the median surviving company, return on assets (as measured by profits before tax in percent of total assets) fell from about 3 percent in 2006/07 to just about 1 percent in Italy and ½ percent in Spain (Figure 4). Profit margins (defined as profit before tax over total operating revenue) generally followed the trend in return on assets (Figure 5). In terms of broad industrial sectors, construction and services were hit the hardest in both countries. The decline in profitability was even deeper in Spain.10 The firm level data reveals remarkable dispersion in profitability across firms, especially within the set of small businesses. In both Spain and Italy, the change in profitability was much more pronounced for the average than the median firm within this category (results available upon request), suggesting that firms at the bottom of the profitability distribution to begin with were hit especially hard. National accounts data for 2014 suggest the decline in profitability (defined as gross operating income in percent of gross value added) moderated significantly in Italy and improved in Spain.
Firm Profitability: Return on Assets
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations. Return on assets is measured as the profit or loss before tax divided by total assets of the firm. The sample is restricted to surviving firms over the 2006-2013 period. The top and bottom 5th percentile of values for ROA are excluded to avoid distortions from outliers.Firm Profitability: Return on Assets
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations. Return on assets is measured as the profit or loss before tax divided by total assets of the firm. The sample is restricted to surviving firms over the 2006-2013 period. The top and bottom 5th percentile of values for ROA are excluded to avoid distortions from outliers.Firm Profitability: Return on Assets
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations. Return on assets is measured as the profit or loss before tax divided by total assets of the firm. The sample is restricted to surviving firms over the 2006-2013 period. The top and bottom 5th percentile of values for ROA are excluded to avoid distortions from outliers.
Firm Profitability: Profit Margin
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations. Profit margin is measured as the profit or loss before tax divided by total operating revenue of the firm. The sample is restricted to surviving firms over the 2006-2013 period. The top and bottom 5th percentile of values for profit margin are excluded to avoid distortions from outliers.Firm Profitability: Profit Margin
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations. Profit margin is measured as the profit or loss before tax divided by total operating revenue of the firm. The sample is restricted to surviving firms over the 2006-2013 period. The top and bottom 5th percentile of values for profit margin are excluded to avoid distortions from outliers.Firm Profitability: Profit Margin
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations. Profit margin is measured as the profit or loss before tax divided by total operating revenue of the firm. The sample is restricted to surviving firms over the 2006-2013 period. The top and bottom 5th percentile of values for profit margin are excluded to avoid distortions from outliers.Leverage
10. Deleveraging has been relatively slow in Italy in light of the stronger balance sheets of Italy’s corporate sector at the onset of the crisis. The leverage of the median and average surviving firm in 2006–13, as captured in the financial debt to assets ratio, peaked in 2012 and declined slightly in 2013 (Figure 6a). 11 The dynamics across firms of different sizes, however, have been quite different. Small and medium firms have made little progress in reducing their leverage and are the most leveraged by the end of the study period. The leverage of the largest firms has continuously, albeit very modestly, declined, but edged up in 2013.12 The national accounts indicate that leverage fell very modestly in 2014. About a fifth of micro firms on the other hand had zero financial debt at the onset of the crisis (consistent with potential barriers to access to finance for these firms). These firms saw a slight increase in financial leverage until 2012 with some signs of deleveraging in the last year for which data are available. The evolution of leverage also varies depending on the initial level of leverage. Firms that had financial debt in 2007 but were levered below the median within their industry have seen a sizable increase in their debt-to-asset ratio (from 0.10 to 0.17 for the median firm) despite the slowdown in aggregate credit growth in Italy. Highly levered firms (those with leverage above the median for their industry in 2007) have kept their financial obligations in line with their assets with a decline of the debt-to-asset ratio from its peak of 0.44 in 2007 for the median company to 0.40 by 2013.
Leverage (Debt-to-Assets)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Leverage (Debt-to-Assets)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Leverage (Debt-to-Assets)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.11. And debt overhang in Italy has increased. Despite efforts by medium and large enterprises to keep leverage in check, financial debt as a share of earnings before taxes, interest, and depreciation (EBITDA) has risen quite sharply since 2007 (by 34 and 48 percent for the average and median firm respectively), reflecting the large reduction in income (part of which reflects cyclical developments). Consistent with the trends in debt-to-assets and profitability, the increase has been the smallest for the largest firms, which were more indebted to begin with (Figure 6b). Debt as a share of EBITDA is higher than in Spain, but the ratio to assets was very similar in 2013.
Debt Overhang (Debt-to-Income)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Debt Overhang (Debt-to-Income)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Debt Overhang (Debt-to-Income)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.12. After peaking at a very high level in 2007, overall leverage has declined much faster in Spain (Figure 6a). National account data suggest the trend continued in 2014. The leverage profile is broadly similar across small and medium sized firms, but even more pronounced among micro firms—perhaps due to the predominance of securitized lending (e.g., individual mortgages), which accelerated during the final phase of the boom. The leverage profile of large firms, however, is relatively less pronounced. In terms of sectoral composition, leverage is the highest in construction, services, and utilities. Despite the rapid pace of deleveraging since the crisis, it took longer for the upward movement in debt-to-income to reverse, and a significant fraction of corporates continues to show symptoms of debt overhang, with financial debt in multiples of income Figure 6b). In trade and market services, the debt-income ratio even continued to increase. Conversely, micro firms reduced debt relative to income at a faster clip than larger enterprises.
Interest Coverage Ratio
13. The fall in profits and slow reduction in debt have curtailed firms’ ability to service their obligations in Italy. The sharp cuts in policy rates in response to the crisis were partially transmitted to firms. Effective interest rates fell by close to 200 bps since 2007 (Figure 7) and aggregated banking system data indicate that rates declined further in 2014 and 2015. However, the interest coverage ratio (ICR)—measured as the earnings before interest, tax, depreciation, and amortization (EBITDA) over interest payments—fell by about 30 percent for the median firm in our sample of surviving enterprises as a result of the contraction in sales and profits (Figure 8).13, 14 The share of companies whose earnings before interest, tax, depreciation, and amortization are insufficient to cover their interest payments (i.e., for which the ICR is less than 1) was roughly 20 percent in 2013. Considering a more widely used threshold of two, the share of vulnerable firms within the sample of surviving firms was about 30 percent.15 Micro firms again saw the biggest decrease in ICRs. However, due to their lower levels of indebtedness at the onset of the crisis, they are not much more likely to have an ICR less than one. The incidence of low ICRs in 2013 was greatest among firms in the construction sector and in market services.
Effective Interest Rates
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Effective Interest Rates
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Effective Interest Rates
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.
Interest Coverage Ratio
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Interest Coverage Ratio
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Interest Coverage Ratio
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.14. In Spain, interest coverage ratios fell through 2012, but have started to improve since 2013. Effective interest rates fell by about 160 basis points since the peak of the crisis, and similar to Italy, continue to decline. However, the fraction of firms struggling to generate enough profits to meet interest payments (i.e., those with ICR<1) increased steadily between 2008 and 2012, from 13 to about 31 percent of surviving firms in 2012, before dropping slightly to 30 percent in 2013 (see also Mendez and Mendez, 2013). For 2014, Mulino and Menendez (2015) document a further slight improvement in ICRs. The pattern is similar when considering an ICR threshold of two. These difficulties have been relatively more pronounced for smaller companies with less than 50 employees, with nearly a half of them having difficulties meeting their interest payments. Among other things, this reflects the fact that the interest rate decline was less pronounced for smaller than for larger firms. Another factor is that, on average, nonfinancial corporates in Spain continue to face relatively high interest rates with spreads still exceeding their precrisis levels.
Vulnerability
15. In both Italy and Spain, corporates remain vulnerable to adverse shocks to profits and interest rates. The share of vulnerable firms increases moderately for a range of shocks to profits and interest rates perhaps reflecting the weak baseline, and a combined shock could raise that proportion by more (Figure 9). Specifically, as illustrative examples, we consider the impact of a 10 percent decline in profits, a 100 bps increase in interest rates and the combination of the two, on the full set of companies for which the necessary data are available in 2013.16 The shock to profits raises the proportion of vulnerable firms (those with ICR<2) by about 2 percentage points in Italy and Spain. Interest coverage ratios appear to be slightly more sensitive to a shock of 100 bps interest rate increase than to the 10 percent reduction in profits. A 100 bps increase in interest rate would increase the share of vulnerable firms by about 3.5 percentage points in Italy, and 4.4 percentage points in Spain. A combination of the two shocks would lead to a 5.5 percentage point rise in vulnerable companies in Italy and 6.6 percent in Spain.
Firm Vulnerability and Banking System Implications
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Firm Vulnerability and Banking System Implications
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Firm Vulnerability and Banking System Implications
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.16. Such adverse shocks could have implications for asset quality of the NFC portfolio in the banking sector. In 2013, the financial debt held by companies with ICR less than one was 22 percent of total corporate sector financial debt (as reflected in Orbis) in Italy and 35 percent in Spain, and debt held by companies with ICR less than two was 34 percent in Italy and 47.5 percent in Spain.17 Under a shock to profits described above, debt at risk (debt of firms with ICR<2) rises by about 3 percentage points in both countries. If interest rates rise, this ratio rises to 42 percent in Italy and 55 percent in Spain. A combined shock could increase the proportion of debt at risk by roughly 10 percentage points in Italy and Spain.
17. Positive shocks to corporate profits, on the other hand, do little to improve the share of debt at risk. A 10 percent positive shock to profits would lower the share of firms with ICR<2 by about 1.4 percentage points and debt of vulnerable companies by 2.4 percentage points in Italy. Similarly, a positive 10 percent shock to profits would lower the share of vulnerable companies and the share of debt at risk by about 1.6 and 2 percentage points in Spain, respectively.18 It is important to keep in mind that a positive shock to overall activity might have a differential impact along the distribution of firms as suggested in recent analyses by both Bank of Italy and Bank of Spain. In particular, the Bank of Spain notes that the recent recovery might have benefited more strongly companies in a more vulnerable situation.
D. Corporate Health, Credit Conditions, and Investment
18. In this environment, firms have cut back investment significantly.19 Despite some variations, all industrial sectors and broad geographical regions experienced strong declines in investment. (Figure 10). The widespread collapse implies that common shocks—such as a fall in aggregate demand, for example—are likely the main driver behind the weak investment. Indeed, a number of papers document that weakness of economic activity is the overriding factor holding back investment in many advanced economies (see Chapter 4 of the April 2015 WEO, Barkbu and others (2015)). These studies estimate that in Italy, about 90 percent of the decline in investment could be explained by the evolution of aggregate output, suggesting that other factors such as financial sector deleveraging, impaired corporate sector balance sheets and debt overhang may also play a role in explaining investment weakness. Similar factors were at work in Spain, where the investment impact of the sudden shortfall in demand seems to have been amplified by the very high precrisis debt levels compared to peers (see Lopez-Murphy, 2014).
Median Change in Investment (2007–13)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Median Change in Investment (2007–13)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.Median Change in Investment (2007–13)
Citation: IMF Staff Country Reports 2015, 167; 10.5089/9781513559681.002.A002
Sources: Orbis and IMF Staff calculations.19. Has the deterioration in the health of the corporate sector contributed to the weakness in investment? In a world of complete markets, firm’s investment would depend solely on the risk-adjusted expected return of a project. In the presence of financial frictions, however, firm’s financial health could play a large role in determining its investment decisions (see Martinez-Carrascal and Ferrando, 2008 for a review of the literature). High leverage may reduce the credit-worthiness of the firm due to asymmetric information between borrowers and lenders. This could raise the borrowing costs it faces or restrict credit supply, particularly during downturns when credit institutions’ risk aversion may rise and for smaller firms where information asymmetries are more pronounced.20 High debt overhang (as captured in the debt to income ratio) may also reduce firm’s incentives to invest as most of the benefits of the new investment would accrue to its debtors (Myers, 1977). Numerous studies, initiated by the Fazzari, Hubbard, and Petersen (1988) have found firm’s balance sheet strength, profitability and the availability of liquid assets to be a significant determinant of investment (see Kalemi-Ozcan and others, 2015 for recent firm-level evidence from Europe).
20. Econometric analysis confirms that in both Italy and Spain weak balance sheets explain part of the weakness in investment. For each country, we estimate a standard firm level investment model for the period 2006–13 with all companies included in the Orbis database, in which the net (and gross) investment rate is a function of firm leverage (measured as debt to assets), debt overhang (captured as EBITDA over debt),21 sales growth, return on assets, and cash-to-assets ratio. Namely,
The return on assets proxies the profitability of the firm. Growth in sales captures firm-specific demand shocks and proxies for future growth opportunities in the absence of measures of future sales expectations (i.e., beyond fluctuations in aggregate demand). The year fixed effects directly control for fluctuations in aggregate demand, overall financial conditions, uncertainty and all other factors that may affect investment equally across firms.22 Firm level fixed effects, on the other hand, absorb all time-invariant heterogeneity across firms.
21. High leverage and debt overhang constrain firm investment in both countries.
Firms cut back investment as leverage (debt-to-assets) and debt overhang increase (i.e., lower EBITDA-to-debt ratio), suggesting the presence of financial constraints (Table 1). These constraints appear to be more binding in Spain than in Italy, with the estimated coefficient on debt-assets more than 1.5 times larger in Spain than in the sample of Italian firms.
Debt-overhang also puts a stronger break on investment in Spain—the estimated coefficient on EBITDA/debt is more than twice the magnitude than in Italy. Several factors might explain the higher sensitivity of investment to balance sheet strength and debt overhang in Spain. First, given the higher indebtedness of Spanish firms, it might reflect nonlinearities in the relationship between investment and debt burden: previous empirical studies have found asymmetric effects of leverage on investment, with balance sheet weakness having a larger impact on investment beyond a certain threshold of leverage/indebtedness (see e.g., Hernando and Martinez-Carrascal, 2008; Jaeger, 2003; Goretti and Souto, 2013; and Ceccheti and others, 2011 for macrolevel evidence). Second, it might signal more restrictive bank lending policies in Spain, especially towards the end of the sample period following the concerted effort to cleanse bank balance sheets with strict provisioning requirements and forbearance criteria, and reclassification of assets on the back of a national Asset Quality Review (AQR). Finally, with regard to the role of debt-overhang, firms with particularly high debt levels might have been less willing to invest in Spain anticipating a larger fraction of their return might go to their creditors. Based on the World Bank Doing Business Survey, creditor rights and the ability to enforce contracts are relatively high in Spain.
Determinants of Net Investment: The Role of Firm Balance Sheets
Determinants of Net Investment: The Role of Firm Balance Sheets
| Italy | Spain | |||||
|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | |
| Debt-Assets | −0.271 *** | −0.236 *** | −0.252 *** | −0.4436*** | −0.4354*** | −0.4330*** |
| [0.008] | [0.010] | [0.010] | [0.011] | [0.015] | [0.015] | |
| EBITDA-Debt | 0.002 *** | 0.002 *** | 0.002 *** | 0.0059*** | 0.0059*** | 0.0041*** |
| [0.000] | [0.000] | [0.000] | [0.000] | [0.000] | [0.000] | |
| Sales Growth | 0.055 *** | 0.055 *** | 0.065 *** | 0.0710*** | 0.0710*** | 0.0728*** |
| [0.002] | [0.002] | [0.004] | [0.002] | [0.002] | [0.005] | |
| Return on Assets | 0.112 *** | 0.111 *** | 0.090 *** | −0.1007*** | −0.1006*** | 0.1715*** |
| [0.010] | [0.010] | [0.018] | [0.012] | [0.012] | [0.0307] | |
| Cash-Assets | 0.777 *** | 0.777 *** | 0.658 *** | 0.7222*** | 0.7224*** | 0.8075*** |
| [0.014] | [0.014] | [0.019] | [0.019] | [0.019] | [0.028] | |
| Debt-Assets * Post | −0.056 *** | −0.031 *** | −0.0109 | −0.0341** | ||
| [0.009] | [0.009] | [0.013] | [0.013] | |||
| EBITDA-Debt * Post | 0 | 0 | 0 | 0.0029** | ||
| [0.000] | [0.000] | [0.000] | [0.000] | |||
| Sales Growth * Post | −0.015 *** | −0.0017 | ||||
| [0.005] | [0.006] | |||||
| Return on Assets * Post | 0.034 * | −0.3582*** | ||||
| [0.019] | [0.032] | |||||
| Cash-Assets * Post | 0.179 *** | −0.1397*** | ||||
| [0.018] | [0.028] | |||||
| r2 | 0.26 | 0.26 | 0.26 | 0.3825 | 0.352 | 0.352 |
| N | 2,998,090 | 2,998,090 | 2,998,090 | 1,615,829 | 1,615,829 | 1,615,829 |
Determinants of Net Investment: The Role of Firm Balance Sheets
| Italy | Spain | |||||
|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | |
| Debt-Assets | −0.271 *** | −0.236 *** | −0.252 *** | −0.4436*** | −0.4354*** | −0.4330*** |
| [0.008] | [0.010] | [0.010] | [0.011] | [0.015] | [0.015] | |
| EBITDA-Debt | 0.002 *** | 0.002 *** | 0.002 *** | 0.0059*** | 0.0059*** | 0.0041*** |
| [0.000] | [0.000] | [0.000] | [0.000] | [0.000] | [0.000] | |
| Sales Growth | 0.055 *** | 0.055 *** | 0.065 *** | 0.0710*** | 0.0710*** | 0.0728*** |
| [0.002] | [0.002] | [0.004] | [0.002] | [0.002] | [0.005] | |
| Return on Assets | 0.112 *** | 0.111 *** | 0.090 *** | −0.1007*** | −0.1006*** | 0.1715*** |
| [0.010] | [0.010] | [0.018] | [0.012] | [0.012] | [0.0307] | |
| Cash-Assets | 0.777 *** | 0.777 *** | 0.658 *** | 0.7222*** | 0.7224*** | 0.8075*** |
| [0.014] | [0.014] | [0.019] | [0.019] | [0.019] | [0.028] | |
| Debt-Assets * Post | −0.056 *** | −0.031 *** | −0.0109 | −0.0341** | ||
| [0.009] | [0.009] | [0.013] | [0.013] | |||
| EBITDA-Debt * Post | 0 | 0 | 0 | 0.0029** | ||
| [0.000] | [0.000] | [0.000] | [0.000] | |||
| Sales Growth * Post | −0.015 *** | −0.0017 | ||||
| [0.005] | [0.006] | |||||
| Return on Assets * Post | 0.034 * | −0.3582*** | ||||
| [0.019] | [0.032] | |||||
| Cash-Assets * Post | 0.179 *** | −0.1397*** | ||||
| [0.018] | [0.028] | |||||
| r2 | 0.26 | 0.26 | 0.26 | 0.3825 | 0.352 | 0.352 |
| N | 2,998,090 | 2,998,090 | 2,998,090 | 1,615,829 | 1,615,829 | 1,615,829 |
22. Liquidity and sales growth are important determinants of investment as well. In both countries, strong growth in sales in the previous period is associated with higher investment, as found in the literature. Investment is also quite sensitive to the availability of cash on hand. The evidence on profitability is mixed. In Italy, higher profitability is associated with higher investment, but the estimated coefficient is negative in Spain.
23. The role of financial constraints has intensified since the crisis. In columns (2), (3), (5) and (6) of Table 1, we examine whether financial constraints have become more binding in the post crisis period, by interacting our measures of debt overhang and leverage with a post 2009 indicator. As expected, the sensitivity of firm investment to net worth rises, suggesting that high leverage puts an even stronger brake on investment in the post-crisis period. Interestingly, it seems that Italian firms have become slightly less responsive to growth opportunities in the post crisis period, perhaps consistent with heightened uncertainty about the future.
24. Large firms are less financially constrained during normal times, though constraints have tightened in the post crisis period even for large firms. We estimate the investment model described above separately for the set of micro, SME, and large firms. As presented in Table 2, and consistent with theory, the financial position of the firm seems to play a much smaller role in driving investment for large firms. In the post crisis period, however, the sensitivity of investment to leverage rises substantially even for the large firms.
Determinants of Net Investment by Firm Size
Determinants of Net Investment by Firm Size
| Italy | Spain | ||||||
|---|---|---|---|---|---|---|---|
| Micro | SME | Large | Micro | Small | Medium | Large | |
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | |
| Debt-Assets | −0.294 *** | −0.159 *** | −0.054 | −0.4665*** | −0.5633 *** | −0.218*** | −0.3082 *** |
| [0.011] | [0.020] | [0.083] | [0.011] | [0.034] | [0.070] | [0.076] | |
| EBITDA-Debt | 0.002 *** | 0.004 *** | 0.008 | 0.0064*** | 0.0025 ** | 0.003 | 0.006 |
| [0.000] | [0.001] | [0.005] | [0.001] | [0.001] | [0.002] | [0.000] | |
| Sales Growth | 0.068 *** | 0.049 *** | 0.011 | 0.0869 *** | 0.0986 *** | 0.0484** | 0.0865 *** |
| [0.005] | [0.008] | [0.032] | [0.007] | [0.012] | [0.027] | [0.022] | |
| Return on Assets | 0.055 *** | 0.281 *** | 0.231 | 0.0926 | 0.2965 *** | 0.4788*** | 0.219 |
| [0.019] | [0.048] | [0.186] | [0.044] | [0.0606] | [0.127] | [0.146] | |
| Cash-Assets | 0.632 *** | 0.880 *** | 0.693 *** | 0.8634 *** | 0.7309 *** | 0.756 *** | 0.6822 *** |
| [0.022] | [0.045] | [0.203] | [0.042] | [0.049] | [0.113] | [0.141] | |
| Debt-Assets * Post | −0.007 | −0.051 *** | −0.191 *** | 0.0201 | −0.0773*** | −0.2717 ** | * −0.0775 |
| [0.011] | [0.017] | [0.064] | [0.019] | [0.028] | [0.057] | [0.068] | |
| EBITDA-Debt * Post | 0 | −0.001 | −0.008 * | 0.0003 | 0.004*** | 0.0025 | 0.000 |
| [0.001] | [0.001] | [0.005] | [0.001] | [0.001] | [0.002] | [0.004] | |
| Sales Growth * Post | −0.011 * | −0.019 ** | −0.022 | −0.027 | −0.0772 *** | 0.003 | −0.049 ** |
| [0.006] | [0.009] | [0.038] | [0.008] | [0.028] | [0.032] | [0.023] | |
| Return on Assets * Post | 0.033 | 0.112 ** | 0.034 | −0.3064*** | −0.5302 *** | −0.6337*** | −0.3462** |
| [0.021] | [0.048] | [0.171] | [0.046] | [0.063] | [0.131] | [0.147] | |
| Cash-Assets * Post | 0.155 *** | 0.169 *** | −0.092 | −0.1545*** | −0.0838 * | −0.1916 | −0.2327 |
| [0.021] | [0.041] | [0.166] | [0.042] | [0.050] | [0.1273] | [0.153] | |
| r2 | 0.27 | 0.22 | 0.21 | 0.4035 | 0.4128 | 0.3955 | 0.6122 |
| N | 2,170,628 | 778,187 | 49,275 | 965,613 | 411,058 | 74,115 | 165,044 |
Determinants of Net Investment by Firm Size
| Italy | Spain | ||||||
|---|---|---|---|---|---|---|---|
| Micro | SME | Large | Micro | Small | Medium | Large | |
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | |
| Debt-Assets | −0.294 *** | −0.159 *** | −0.054 | −0.4665*** | −0.5633 *** | −0.218*** | −0.3082 *** |
| [0.011] | [0.020] | [0.083] | [0.011] | [0.034] | [0.070] | [0.076] | |
| EBITDA-Debt | 0.002 *** | 0.004 *** | 0.008 | 0.0064*** | 0.0025 ** | 0.003 | 0.006 |
| [0.000] | [0.001] | [0.005] | [0.001] | [0.001] | [0.002] | [0.000] | |
| Sales Growth | 0.068 *** | 0.049 *** | 0.011 | 0.0869 *** | 0.0986 *** | 0.0484** | 0.0865 *** |
| [0.005] | [0.008] | [0.032] | [0.007] | [0.012] | [0.027] | [0.022] | |
| Return on Assets | 0.055 *** | 0.281 *** | 0.231 | 0.0926 | 0.2965 *** | 0.4788*** | 0.219 |
| [0.019] | [0.048] | [0.186] | [0.044] | [0.0606] | [0.127] | [0.146] | |
| Cash-Assets | 0.632 *** | 0.880 *** | 0.693 *** | 0.8634 *** | 0.7309 *** | 0.756 *** | 0.6822 *** |
| [0.022] | [0.045] | [0.203] | [0.042] | [0.049] | [0.113] | [0.141] | |
| Debt-Assets * Post | −0.007 | −0.051 *** | −0.191 *** | 0.0201 | −0.0773*** | −0.2717 ** | * −0.0775 |
| [0.011] | [0.017] | [0.064] | [0.019] | [0.028] | [0.057] | [0.068] | |
| EBITDA-Debt * Post | 0 | −0.001 | −0.008 * | 0.0003 | 0.004*** | 0.0025 | 0.000 |
| [0.001] | [0.001] | [0.005] | [0.001] | [0.001] | [0.002] | [0.004] | |
| Sales Growth * Post | −0.011 * | −0.019 ** | −0.022 | −0.027 | −0.0772 *** | 0.003 | −0.049 ** |
| [0.006] | [0.009] | [0.038] | [0.008] | [0.028] | [0.032] | [0.023] | |
| Return on Assets * Post | 0.033 | 0.112 ** | 0.034 | −0.3064*** | −0.5302 *** | −0.6337*** | −0.3462** |
| [0.021] | [0.048] | [0.171] | [0.046] | [0.063] | [0.131] | [0.147] | |
| Cash-Assets * Post | 0.155 *** | 0.169 *** | −0.092 | −0.1545*** | −0.0838 * | −0.1916 | −0.2327 |
| [0.021] | [0.041] | [0.166] | [0.042] | [0.050] | [0.1273] | [0.153] | |
| r2 | 0.27 | 0.22 | 0.21 | 0.4035 | 0.4128 | 0.3955 | 0.6122 |
| N | 2,170,628 | 778,187 | 49,275 | 965,613 | 411,058 | 74,115 | 165,044 |
25. Tight credit supply may have also contributed to investment weakness both in Italy and Spain. So far, we have documented the importance of financial frictions operating via firms: as firms’ financial health deteriorates, they become risky borrowers. But weak bank balance sheets may prevent banks from lending to any borrower. To test this hypothesis, we examine whether firms in sectors more dependent on external finance cut investment less in Italian regions where growth in credit outstanding fell less.23 Table 3 suggests that this was indeed the case. Since credit outstanding captures both demand and supply of credit, we explore a more direct (albeit still imperfect) measure of credit supply: lending standards reported by banks, which are available for the four Italian macro-regions, and separately for firms in manufacturing, construction and services since 2009.24 Using this alternative measure of credit supply by banks, we do find that firms in sectors more dependent on external finance cut investment more in areas and sectors where lending standards were tightened relatively more.25 This broadly matches results reported by Jiménez and others (2015) for Spain. Matching company and loan application data between 2002 and 2010, they report that firm leverage is a strong predictor of the approval of loan applications in both good and crisis times. In contrast, bank balance-sheet strength determines the granting of loan applications only during crisis times, suggesting that weak bank balance sheets also weighed on investment in Spain.
Net Investment in Italy: The Role of Firm Balance Sheets and Credit Conditions
Net Investment in Italy: The Role of Firm Balance Sheets and Credit Conditions
| Net Investment | ||||
|---|---|---|---|---|
| (1) | (2) | (3) | ||
| Estimation Period | 2006-13 | 2009-13 | 2010-13 | |
| Debt-Assets | −0.271 *** | −0.345 *** | −0.409 *** | |
| [0.008] | [0.013] | [0.017] | ||
| EBITDA-Debt | 0.002 *** | 0.002 *** | 0.003 *** | |
| [0.000] | [0.000] | [0.001] | ||
| Sales Growth | 0.055 *** | 0.042 *** | 0.031 *** | |
| [0.002] | [0.003] | [0.004] | ||
| Return on Assets | 0.112 *** | 0.045 *** | 0.054 *** | |
| [0.010] | [0.016] | [0.020] | ||
| Cash-Assets | 0.777 *** | 0.768 *** | 0.794 *** | |
| [0.014] | [0.023] | [0.028] | ||
| Financial Dependence * Credit Growth | 0.059 * | |||
| [0.030] | ||||
| Financial Dependence * Lending Standards | −0.085 *** | |||
| [0.031] | ||||
| r2 | 0.26 | 0.34 | 0.39 | |
| N | 2,998,090 | 1,600,368 | 1,279,203 | |
Net Investment in Italy: The Role of Firm Balance Sheets and Credit Conditions
| Net Investment | ||||
|---|---|---|---|---|
| (1) | (2) | (3) | ||
| Estimation Period | 2006-13 | 2009-13 | 2010-13 | |
| Debt-Assets | −0.271 *** | −0.345 *** | −0.409 *** | |
| [0.008] | [0.013] | [0.017] | ||
| EBITDA-Debt | 0.002 *** | 0.002 *** | 0.003 *** | |
| [0.000] | [0.000] | [0.001] | ||
| Sales Growth | 0.055 *** | 0.042 *** | 0.031 *** | |
| [0.002] | [0.003] | [0.004] | ||
| Return on Assets | 0.112 *** | 0.045 *** | 0.054 *** | |
| [0.010] | [0.016] | [0.020] | ||
| Cash-Assets | 0.777 *** | 0.768 *** | 0.794 *** | |
| [0.014] | [0.023] | [0.028] | ||
| Financial Dependence * Credit Growth | 0.059 * | |||
| [0.030] | ||||
| Financial Dependence * Lending Standards | −0.085 *** | |||
| [0.031] | ||||
| r2 | 0.26 | 0.34 | 0.39 | |
| N | 2,998,090 | 1,600,368 | 1,279,203 | |
26. The implications of the declining health of the corporate sector for aggregate business investment are non-trivial. A back of the envelope calculation using the point estimates in Table 1 and the change in net worth, debt overhang, sales, and profitability over 2006-13, gives a rough sense of how much of the decline in aggregate corporate investment is due to firms’ deteriorating health. We find that this deterioration can explain about 8–10 percent of the decline in aggregate investment in Italy. Using aggregate data, Busetti, Giordano, and Zevi (2015) find that up to one third of the decline in nonconstruction investment in Italy was due to credit supply constraints in the most acute phases of the recession.
E. Conclusion
27. The crisis has had a severe impact on the corporate sector in Italy and Spain. The analysis of a rich firm-level dataset up to 2013 highlights a number of stylized facts.
Firms in both countries are indebted and suffer from weak profitability, recent improvements notwithstanding. As a result, their capacity to deal with shocks is limited, with important potential implications for the banking sector.
Starting from a higher initial level, leverage has come down significantly in Spain, driven by a remarkable reduction in net lending starting in 2008, when firms became net savers, and since 2013, has been supported by increasing profits and strengthening recovery. Nevertheless, leverage is still high relative to peers and historical standards. Starting from a better initial position, corporate deleveraging has been much slower in Italy, and debt overhang has actually risen due to the sharp fall in firm income.
Weak balance sheets and debt overhang are putting a break on business investment, thus slowing the process of recovery in both economies. These effects have been stronger in Spain during the crisis, where they have been compounded by the impact of the deep restructuring and strengthening of banks’ balance sheets under the ESM-led program. More recently, bank balance sheets and profitability have considerably improved thanks to those measures and business investment has picked up strongly. In Italy, there is also evidence that tight credit supply may have also contributed to investment weakness.
28. These findings highlight the need for a broad strategy to repair and revive the corporate sector. This strategy should aim to ease the reallocation of resources towards the healthier firms and sectors in the economy, rehabilitate the financial and operational structure of viable enterprises to give them a fresh start, and expand firms’ access to alternative sources of funds to strengthen their resilience and reduce the interdependence of the corporate and banking sectors in the future.
References
Barkbu, Bergljot, S. Pelin Berkmen, Pavel Lukyantsau, Sergejs Saksonovs, and Hanni Schoelermann, 2015, “Investment in the Euro Area: Why Has it Been Weak,” IMF Working Paper No. 15/32 (Washington: International Monetary Fund).
Bontempi, M. E., R. Golinelli and G. Parigi, 2010, “Why Demand Uncertainty Curbs Investment: Evidence from a Panel of Italian Manufacturing Firms.” Journal of Macroeconomics, Vol. 32, pp. 218–38.
Busetti, F., C. Giordano, and Giordano Zevi, 2015, “Main Drivers of the Recent Decline in Italy’s Nonconstruction Investment.” Bank of Italy Occasional Paper, forthcoming.
CERVED, 2014, “2014 Cerved SME Report.”
Cecchetti, S.G., M.S. Mohanty, and F. Zampolli, 2011, “The Real Effects of Debt,” BIS Working Papers 352 (Bank for International Settlements).
Claessens, Stijn, Hui Tong and Shang-Jin Wei, 2012, “From the Financial Crisis to the Real Economy: Using Firm-Level Data to Identify Transmission Channels,” Journal of International Economics, Vol. 88, pp. 375–387.
European Central Bank, 2014, “Deleveraging Patterns in the Euro Area Corporate Sector,” ECB Monthly Bulletin (February).
European Commission, 2014, “A Partial and Fragile Recovery,” Annual Report on European SMEs 2013/20014 (July).
European Commission, 2014, SME Performance Review Database. Available at: http://ec.europa.eu/enterprise/policies/sme/facts-figures-analysis/performance-review/index_en.htm
European Commission, 2015, Country Report Italy 2015 (February).
Fazzari, Steven, Glenn Hubbard, and Bruce Petersen, 1988, “Financing Constraints and Corporate Investment,” Brookings Papers on Economic Activity, Vol. 1 pp. 141–95.
Gertler, Mark and Simon Gilchrist, 1994, “Monetary Policy, Business Cycles, and the Behavior of Small Manufacturing Firms,” The Quarterly Journal of Economics Vol. 59, pp. 309–40.
Gertler, Mark and Simon Gilchrist, 1995, “The Role of Credit Market Imperfections in the Monetary Transmission: Arguments and Evidence,” Scandinavian Journal of Economics Vol. 95, pp. 43–64.
Goretti, Manuela and Marcos Souto, 2013, “Macro-Financial Implications of Corporate (De)Leveraging in the Euro Area Periphery,” IMF Working Paper WP/13/154 (Washington: International Monetary Fund).
Hernando, Ignacio and Martinez-Carrascal, 2008, “The Impact of Financial Variables on Firm’s Real Decisions: Evidence from Spanish Firm-Level Data,” Journal of Macroeconomics, Vol. 30, pp. 543–561.
International Monetary Fund, 2013, “Italy 2013 Article IV Consultation,” IMF Country Report No. 13/298 (Washington).
International Monetary Fund, 2014, “Tackling the Corporate Debt Overhang in Spain,” Spain—Selected Issues (Washington).
International Monetary Fund, 2015, “Resolving Nonperforming Loans in Italy: A Comprehensive Approach,” Italy—Selected Issues (Washington).
International Monetary Fund, 2015, “Private Investment: What’s the Holdup?” World Economic Outlook: Uneven Growth—Short- and Long-Term Factors, April (Washington).
ISTAT, 2014, “Struttura e Competitivita del Sistema delle Imprese Industriali e dei Servizi, Anno 2012,” Statistiche Report (November 27).
Jaeger, A., 2003, “Corporate Balance Sheet Restructuring and Investment in the Euro Area,” IMF Working Paper WP/03/117 (Washington: International Monetary Fund).
Jiménez, Gabriel, Steven Ongena, José-Luis Peydró, and Jesús Saurina, 2014, “‘Credit Demand Forever?’ On the Strengths of Bank and Firm Balance-Sheet Channels in Good and Crisis Times,” Bank of Spain.
Jones, M. T. and M. Karasulu, 2006, “The Korean Crisis: What Did We Know and When Did We Know It? What Stress Test of the Corporate Sector Reveal,” IMF Working Paper WP/06/114 (Washington: International Monetary Fund.
Kalemli-Özcan, Sebnem, Luc Laeven, and David Moreno, 2015, “Debt Overhang in Europe: Evidence from Firm-Bank-Sovereign Linkages,” unpublished manuscript.
Mendez, M. and A. Menendez, 2013a, “La Evolucion del Endeudamiento de las Empresas No Financieras desde el Inicio de la Crisis,” Economic Bulletin, January (Madrid: Bank of Spain).
Mendez, M. and A. Menendez, 2013b, “Resultados de las Empresas No Financieras en 2012 y Hasta el Tercer Trimestre del 2013,” Economic Bulletin, November (Madrid: Bank of Spain).
Mulino, M. and A. Menendez, 2015, “Resultados de las empresas no financieras. Cuarto trimestre de 2014 y avance de cierre del ejercicio,” Economic Bulletin, March (Madrid: Bank of Spain).
Myers, Stewart, 1977, “Determinants of Corporate Borrowing,” Journal of Financial Economics Vol. 5, No. 2, pp. 147–175.
Orbis, 2014, Orbis Data Guide, Bureau van Dijk.
Rajan, Raguram and Luigi Zingales, 1998, “Financial Dependence and Growth,” American Economic Review, Vol. 88, No. 3, pp. 559–86.
Tong, Hui and Shang-Jin Wei, 2011, “The Composition Matters: Capital Inflows and Liquidity Crunch During a Global Economic Crisis,” Review of Financial Studies, Vol. 24, No. 6, pp. 2023–52.
Vermeulen, Philip, 2002, “Business Fixed Investment: Evidence of a Financial Accelerator in Europe,” Oxford Bulletin of Economics and Statistics, Vol. 64, pp. 213–31.
Prepared by Nina Budina, Sergi Lanau, and Petia Topalova (EUR). We thank Vizhdan Boranova and David Velazquez-Romero for excellent research assistance. This is a cross-country Selected Issues paper and will also serve as background material for the upcoming Executive Board Meeting on Spain 2015 Article IV Consultation.
The sample of firms for which 2014 financial statements were available was too small to conduct a systematic analysis. Where possible, the findings are complemented by a discussion of the corporate sector data available in the 2014 national accounts.
In 2012, more than 99.9 percent of businesses employed fewer than 50 people in Italy and in Spain. These businesses accounted for 70 percent of value added and 54 percent of overall employment in Italy (ISTAT, 2014). In Spain, they comprise slightly less than ½ of value added and about 60 percent of total employment (SME Performance review database, 2014).
An important caveat is that about half of the firms in the sample report their financials (e.g., debt, assets, etc.) due to simplified filing requirements for SMEs (Orbis, 2014).
Jones and Karasulu (2006), for example, found that stress testing applied to the balance sheets of Korean corporations in advance of the Asian crisis of the late 1990s would have shown the degree of vulnerabilities present in the country.
This estimate will present an upper bound since ICR do not capture all the resources companies may have to meet their obligations and to smooth over the idiosyncratic shocks to profits they may experience.
Kalemli-Özcan, Laeven and Moreno (2015) examine the role of debt overhang, leverage, and banking sector weakness in investment using a more detailed version of the Orbis database for all European economies.
More specifically, the downside from this approach is that (i) our analysis does not reflect the churn that might have occurred in response to the downturn, a potentially important margin of adjustment, and (ii) there might be differences in the time trends of the indicators we study relative to the available macro data due to survivors’ bias introduced by the sample selection. Namely, the decline in profitability or interest coverage ratios might be more pronounced within our sample, as firms that had successfully weathered the crisis and the protracted downturn were likely a stronger subset of the universe of firms in Italy and Spain—hence, their profitability and liquidity were likely higher prior to the crisis. That said, by the end of the sample period, the crisis had likely taken a heavy toll on these firms relative to the new entrants in the corporate sector, exacerbating the fall in measured profitability. Only firm census data could accurately capture the time trends in the corporate sector.
There are significant gaps in the coverage of number of employees for Italian firms in Orbis, hence we use firm assets to classify firms into micro, small, medium, and large. However, given this difference in the definitions, some caution is required when comparing directly levels and, to a smaller degree, dynamics of variables by firm size in the two countries. Classifying Spanish firms by asset size does not alter significantly the findings but results in an almost-empty group of large firms.
The broad sectors considered here include manufacturing, construction, utilities (electricity, gas, water supply and sewage), wholesale and retail trade, market services (transport and storage, accommodation, professional and technical, and real estate, entertainment and other services), and basic services (administrative support, education, health).
We define leverage as the sum of short term financial debt (Loans) and long term financial debt (long term loans and other noncurrent liabilities), divided by total assets. Note that this definition of debt excludes provisions, trade credits and other short term liabilities (such as pensions, deferred taxes, accounts receivables in advance) that are part of firms’ total debt. The patterns described above are similar if one uses a more restrictive definition of financial debt to include only long term and short-term loans as in ECB (2014).
Alternative measures of leverage, namely the ratio of debt to debt plus equity, suggest that among firms that had financial debt, leverage was the highest among the smallest companies, and the lowest among the large companies, in line with analysis by the Bank of Italy of a larger sample of firms with substantially more detailed financial data.
Using an alternative measure of earnings, EBIT yields qualitatively similar findings of the overall decline in ICR, and patterns across companies of different sizes and in different industries.
It is important to note the sample of surviving firm likely overstates the decline in ICR and the rise of the share of vulnerable firms over time relative to what actually occurred in the overall corporate sector. Firms that survived through the severe economic downturn were likely a stronger subset of the universe of firms and hence they had a higher ICR at the onset of the crisis.
A threshold of two was used in the 2013 joint IMF-Bank of Italy FSAP. The report notes that an ICR below two is broadly consistent with B ratings or lower by rating agencies, suggestive of about 20 percent probability of default over a 5-year horizon.
For this exercise, in both the Spanish and Italian sample of firms, we exclude firms with zero or negative reported debt in 2013, as well as firms with debt in the top 1 percentile of the distribution.
Note that these estimates are slightly different from debt-at-risk estimates of the Bank of Italy and Bank of Spain. In Italy, debt-at-risk is about 2 percentage points lower than Bank of Italy estimates, while in Spain it is about 5 percentage points higher than Bank of Spain estimates which, as mentioned, also point to improvements more recently. This reflects the smaller sample size, the less precise measurement of bank debt, and differences in data cleaning methodologies.
A large share of firms in both Spain and Italy report negative profits in 2013. This results in little change in the share of vulnerable firms in response to a positive shock modeled as a percent change in profits. To allow for the possibility of firms operating at a loss to swing into positive territory, we examine the consequence of a shock to operating revenue, holding expenses constant. The shock is calibrated to approximate a 10 percent increase in EBITDA for the average firm. As expected, such a shock leads to a more sizable reduction in the share of debt at risk in both countries.
Firm-level data suggest a much deeper decline in investment in 2010 than is discernible in the macro data. We use two different measures of investment. Gross investment is defined as the change in total fixed assets between t and t-1 plus depreciation and amortization over total fixed assets at t-1. Net investment is defined as the change in total tangible fixed assets over the total tangible fixed assets in the previous year.
We follow Kalemi-Ozcan and others (2015) and use the inverse of debt overhang (i.e., EBITDA over debt, rather than the debt-to-income ratio) since earnings may be zero or negative.
Data limitations prevent us from examining the role of firm-specific uncertainty shocks, which have been shown to significantly dampen investment in the case of Italy (see Bontempi, Golinelli, and Parigi, 2010).
A sector’s dependence on external finance is from Tong and Wei (2011), who build on the methodology first developed by Rajan and Zingales (1998). Specifically, financial dependence of a sector is constructed as the difference of the capital expenditures of the sector and its cash flow as a share of its total capital expenditures in the 1990–2006 period in the US.
The measure of credit supply reflects the response of banks to the question on whether they have tightened or relaxed lending standards in the previous three months.
Of course, the change in lending standards may reflect not only the strength of bank’s balance sheets and its willingness to lend but also the perceived riskiness of borrowers.
