Private non-financial sector indebtedness: where do we stand?

Transcription

1 HCSF/217/ e séance Private non-financial sector indebtedness: where do we stand? The French private non-financial sector (households and firms) indebtedness registered a steady increase since the 27 global financial crisis. It reached 129.6% of GDP at the end of the second quarter of 217 1, which corresponds to a 29.6 basis points rise since 28. As of end-june 217, total loans to non-financial corporations (NFCs) reached 71.7% of GDP, while total loans to households amounted 57.9% of GDP. The analysis of additional indicators shows that: Loans to private non-financial sector are beyond their long-term trend. The Basel gap (or the broad credit-to-gdp gap), indicator assessed in a large number of countries, corresponds to the deviation of broad (both bank and market) debt of households and NFCs with respect to its long-term trend. In France, this broad gap is highly positive, being mainly explained by the sharp increase in NFCs issuance of debt securities. The banking credit-to-gdp gap differs from the Basel gap as it focuses on domestic indebtedness vis-à-vis banks 2. In France, the banking gap is on an upward trend and reaches now its highest level since the first quarter of 214, mainly due to the pronounced increase in credit to households and the continued strong indebtedness of NFCs. Credit dynamics largely overpass those of real economy. Broad credit relative to GDP registered a sharp increase and reached +5.6 percentage points in 217 Q2. At the same time, bank credit over GDP increased strongly, with a year-on-year growth of +2.7 percentage points in 217 Q2. This report assesses the developments in the indebtedness of French NFCs. Its main objectives consist in describing the context of debt expansion as well as its implications for the private nonfinancial sector. 1 This figure excludes intragroup non-financial corporation loans. Securities are accounted for in nominal values. With securities at market values, the private non-financial sector debt is of 131.8% of GDP. 2 This stands for debt instruments (bank loans and debt securities) issued by domestic private non-financial sector actors, registered as assets in the balance sheet of domestic banks. 1

2 Q1 1997Q2 1998Q3 1999Q4 21Q1 22Q2 23Q3 24Q4 26Q1 27Q2 28Q3 29Q4 211Q1 212Q2 213Q3 214Q4 216Q1 217Q2 1996Q1 1997Q2 1998Q3 1999Q4 21Q1 22Q2 23Q3 24Q4 26Q1 27Q2 28Q3 29Q4 211Q1 212Q2 213Q3 214Q4 216Q1 217Q2 Figure 1 - Indebtedness of non-financial private and public sectors (in percentage of GDP) Figure 2 - Indebtedness of households and nonfinancial corporates (in percentage of GDP) Public sector Non-financial private sector Households Non financial corporates Source : INSEE, Banque de France. Last observation: 217 Q2. Source : INSEE, Banque de France. Last observation : 217 Q2. Figure 3 Broad Credit-to-GDP gap, France (in percentage) Figure 4 - Bank Credit-to-GDP gap, France (in percentage) 15% 15% 15% 15% 1% 1% 1% 1% 5% 5% 5% 5% % % % % -5% -5% -5% -5% Households contribution NFC contribution Broad gap Households contribution NFC contribution Bank gap Source : BIS, INSEE, Banque de France. Last observation: 217 Q2. Source : BIS, INSEE, Banque de France. Last observation: 217 Q2. Box 1 : The Basel gap advantages and limits (according to Couaillier and Idier, 217, Bulletin de la Banque de France) In economic research there was an increased interest in measuring credit in excess, as a financial crisis forerunner. The consensus that emerged took the form of the «credit to GDP gap» or Basel gap, i.e. the deviation of credit to GDP ratio from its long term trend, computed over a time span starting in 197. Several studies, mainly within the Basel Committee, have shown ex post that this indicator could have predicted several financial crises since 198 and in particular that of 28. Nowadays, this indicator belongs to the list of reference indicators used for the calibration of countercyclical capital buffer. The Basel gap represents a standardized indicator of credit in excess largely used in the euro area and the European Union countries. Its computation, as recommended by the Basel Committee, requests the use of broad credit data, i.e. data covering all the sources of borrowing of the private non-financial sector (Dembiermont et al., 21). In particular, it should include all credit extended to households and other private non-financial entities in an economy, irrespective of its form and/or the identity of the supplier of funds. This means that it should include credit extended by domestic and international banks, as well as non-bank financial institutions either domestically or directly from abroad, and should also include all debt securities issued domestically or internationally to fund households and other private non-financial sector entities (including the securitization, regardless of who holds the securities). 2

3 Despite the interest in having a standardized indicator for the surveillance of credit in excess (and of indebtedness), a robust and accurate diagnosis should be based on an exhaustive set of indicators. The Basel gap presents some limitations in terms of the statistical definition of credit and debt components, as well as some methodological limits with respect to the definition of long term trend of the credit to GDP ratio (cf. Couaillier and Idier, 217). Among the additional indicators we could quote an alternative gap also called bank credit gap. It focusses exclusively on the assets of banks, as opposed to the Basel gap that focusses on the liabilities of private non-financial sector actors. Its definition is narrower and allows a better appraisal, in the diagnosis, of risks that are solely inherent to bank debt. 3

4 1999Q1 2Q1 21Q1 22Q1 23Q1 24Q1 25Q1 26Q1 27Q1 28Q1 29Q1 21Q1 211Q1 212Q1 213Q1 214Q1 215Q1 216Q1 217Q1 Developments in French NFCs debt I. The indebtedness of firms continues to increase, it rises even more sharply for large firms A. The increase in French NFCs debt contrasts with developments in NFCs debt in the euro area The increase in private non-financial sector debt is mainly explained by the outstanding debt of French NFCs that was of roughly 161 billion in 217 Q2, with 11 billion of outstanding bank credit debt 3 and 6 billion of debt securities 4. Since 21, total indebtedness of French NFCs increased at an average yearly growth rate of +4.7%. These dynamics result in a continued increase in indebtedness related to GDP, the added-value and also the turnover of NFCs. The French private non-financial sector debt is close to euro-area levels. However, its evolution contrasts with the relative stabilization or the decrease observed in some large European countries. Since end- 21, the ratio of NFCs debt to GDP declined by 1.7 percentage points in the euro area, while it has increased by 11.5 percentage points in France. As a consequence, in 217 Q2, French NFCs debt over GDP was 7.3 percentage points higher than in the euro area Figure 5 - Gross indebtedness of NFCs (in percentage of GDP) Germany Spain France Eurozone Italy Note: Debt is consolidated; it corresponds to the sum of debt securities issued by NFCs, in nominal value, and of total credit netted of claims registered in the assets. Debt computed this way consolidates intragroup credit vis-à-vis resident and non-resident entities, allowing more appropriate international comparisons (according to Couaillier and Idier, 217, and Appendix in the current note). The consolidated debt should not be mixed up with the net debt, the latter consisting in the difference between gross debt and the cash positions of NFCs. See Appendix (debt consolidated at the level of the institutional sector, 2 nd approach). Source: Banque de France. Last observation: 217 Q2. In 217 Q2, the increase in NFCs bank loans adjusted for securitization was of.6% in the euro area. This figure can be decomposed in a contribution of French NFCs of roughly 1.1% and, respectively, of -.5% for the rest of the euro area. As regards debt securities issued by NFCs, the increase in the outstanding amounts was of 5.4% in the euro area, with a significant contribution of French NFCs debt securities, of roughly 3.1%. 3 When considering only credit extended by French banks, this figure stands at 93 billion. 4 This figure stands for debt securities in nominal values. For debt securities at market value, the outstanding debt is of 66 billion. 4

5 211Q1 211Q3 212Q1 212Q3 213Q1 213Q3 214Q1 214Q3 215Q1 215Q3 216Q1 216Q3 217Q1 Jan-11 May-11 Sep-11 Jan-12 May-12 Sep-12 Jan-13 May-13 Sep-13 Jan-14 May-14 Sep-14 Jan-15 May-15 Sep-15 Jan-16 May-16 Sep-16 Jan-17 May-17 Sep-17 Figure 6 - Growth of bank credit of NFC in the euro area (in percentage) 2% 1% % -1% -2% -3% -4% -5% -6% Contribution of the rest of Eurozone Contribution of France Growth of Eurozone Source: European Central Bank. Last observation: September 217. Note: loans are adjusted for securitization. For the euro area, the figures stand for the growth of credit extended by the euro area monetary and financial institutions to NFCs from the euro area. For France, its contribution corresponds to credit extended by French monetary and financial institutions to French NFCs; it underestimates the share of French NFCs, as it excludes credit extended by euro area banks to French NFCs. Figure 7 - Growth of debt securities issued by NFCs in the euro area (in percentage) 25% 2% 15% 1% 5% % -5% -1% Source: European Central Bank. Last observation: 217 Q2. France contribution Rest of Eurozone contribution Eurozone growth 5

6 Dec-6 Jun-7 Dec-7 Jun-8 Dec-8 Jun-9 Dec-9 Jun-1 Dec-1 Jun-11 Dec-11 Jun-12 Dec-12 Jun-13 Dec-13 Jun-14 Dec-14 Jun-15 Dec-15 Jun-16 Dec-16 Jun-17 B. The dynamics are different depending on firms categories As of end-june 217, total debt of French NFCs 5 consisted of 684 billion debt of large enterprises (LE), 416 billion debt of firms of an intermediary size (ISE) and, respectively, 514 billion debt of small and medium sized enterprises (SME). Figure 8 - Gross debt of non-financial corporates, by firm size (in billion euros) Large enterprises Intermediate sized enterprises SME Note: NFCs debt is consolidated at the level of the institutional sector, 3rd approach (see Appendix). The decomposition is done according to the size for credit in the SCI and others item. Source: Banque de France. Last observation: June 217. Over the period 211 to 216, the evolution of indebtedness was more pronounced for LE and ISE (corresponding to 3 percentage points and, respectively, 2 percentage points of their added-value, namely an average year-on-year growth rate of 7.5% and 6% respectively, cf. Figure 9) compared to SME (an average year-on-year growth rate of 2 percentage points). It should be noted that a large part of LE and ISE debt belong to property holding companies (SCI, Sociétés Civiles Immobilières) 6. 5 Excluding credit extended by foreign banks. In what follows, we retain the concept of debt consolidated at the level of the institutional sector that is measured in accordance with the third approach presented in the Appendix. Loans are taken from the assets of financial institutions. 6 For bank credit, the decomposition is based on the Banque de France Central Division of Risks (Service Central des Risques) data, complemented with the distribution by size of the SCI and others item as published in StatInfo Loans by firms size. For debt securities, the decomposition is based on a matching between the database on the issuance of securities (code ISIN/SIREN issuer) and the central balance sheet database/ ESANE (SIREN characteristics - workforce and perimeter of the group associated with the ISIN code). 6

7 Figure 9 - Average yearly growth in NFCs indebtedness ( , in percentage) Note: NFCs debt is consolidated at the level of the institutional sector, 3rd approach (see Appendix). The decomposition is done according to the size for credit in the SCI and others item. Source: Banque de France. As far as French SME are concerned, the slow increase in their indebtedness seems not to be due to restrictive credit conditions. Indeed, the Banque de France and ECB surveys on SMEs financing indicate both a high level of credit demand and a high credit supply in response to this demand. Thus, in the Banque de France survey, the observed rate of response to the demand of credit (i.e. the percentage of demand of credit that was fulfilled entirely or at least for 75% of their amount) reached its maximum both for cash credit (86%) and investment credit (95%). C. Market financing played a significant role Market debt has significantly increased in recent years, as a result of a swing in the financing of large enterprises towards debt securities. Even though it pertains to a large number of enterprises, the analysis of the issuances of debt securities indicate a strong concentration in the utilities sector, which represents a large issuer. The contributions of bank credit and market financing to the global dynamic of indebtedness have considerably evolved over time.over the period 211 to 217 Q2, the large enterprises registered a strong increase in bond financing (+48%) coupled with a lower rise in bank financing (+9%). The expansion of bond financing was equally pronounced for intermediary size enterprises (+149%); however, the outstanding amounts of debt securities issued by these enterprises remain rather low (at roughly 6 billions). Over the recent years, the dynamic of NFCs debt is explained by both bank credit and debt securities. Since mid-216, the increase of outstanding bank credit remains steady, while market financing speed up. As of end-september 217, the year-on-year growth rate of bank credit was of roughly +5.4% and debt securities registered a year-on-year growth rate of +7.7%. 7

8 S S1 Figure 8 Gross debt of non-financial corporates, by firm size (bank credit, in billions of euros) Figure 9 Gross debt of non-financial corporates, by firm size (debt securities, in billions of euros) Large enterprises Intermediate sized enterprises Large enterprises Intermediate sized enterprises SME Total SME Total Note: NFCs debt is consolidated at the level of the institutional sector, 3rd approach (see Appendix). The decomposition is done according to the size for credit in the SCI and others item. Source : Banque de France. Last observation : 217 S1. II. The decline in interest rates has mitigated the impact of strong indebtedness on debt burden The rise in NFCs debt took place in a context of a sharp drop in interest rates since 211, mainly explained by the monetary policy stance in the euro area and, more generally, by a long lasting low interest rate environment. The average cost of bank credit to NFCs is stable and stands at roughly 1.4% since end-216, its lowest historical level. Bank credit rates have slumped to 1% in 216, and only slightly increased since then. The cost of financing by bonds issuance, which mainly concerns the large enterprises, has followed the same downward trend. It slumped to levels below the bank credit rate in 213 and stabilized at roughly.8% since mid-216. As regards small and medium sized enterprises, the average cost of market financing followed a parallel evolution though at higher levels and settled near 1.7% in 217 Q3. 8

9 Figure 1 Developments in interest rates for loans to NFCs, by firm size (overall effective rate, in percentage points) 6,5 6, 5,5 5, 4,5 4, 3,5 3, 2,5 2, 1,5 1,,5, SME Intermediate sized enterprises Micro-enterprises Large enterprises Source: Banque de France. Last observation: 217 Q3. Given the decline in interest rates (lower on average than in the rest of the euro area), the financial burden compared to gross operating surplus decreased over the last years. However, the decrease in the financial burden of French NFCs should not be overstated. Indeed, it remains at rather high levels (between 1% and 16% of earnings before interest, taxes, depreciation and amortization, EBITDA) because of the steady increase in indebtedness and could be challenged in case of an increase in borrowing rates: (i) a large share of bank debt is at variable rate and would therefore be directly affected by a rise in rates; (ii) unlike households, NFCs are often confronted with situations of debt refinancing, this makes them more sensitive to a rate hike, even on their fixed rate debt. Nevertheless, over the recent period, firms have benefited from a favorable period to extend the maturity of their debt, thus partially offsetting the effects of higher rates. In addition, a gradual increase in interest rates during the recovery of the activity would be offset by rising earnings before interest, taxes, depreciation and amortization (EBITDA) of French corporates, and would limit the rise in debt burden. 9

10 Dec-99 Sep- Jun-1 Mar-2 Dec-2 Sep-3 Jun-4 Mar-5 Dec-5 Sep-6 Jun-7 Mar-8 Dec-8 Sep-9 Jun-1 Mar-11 Dec-11 Sep-12 Jun-13 Mar-14 Dec-14 Sep-15 Jun-16 Mar-17 Figure 11 The share of interest expense in gross operating surplus (%) SME 3-large enterprises 2-Intermediate sized enterprises AA-Total Source: Banque de France FIBEN data. Last observation: 216. Note: the share of debt measure includes interest on unconsolidated debt, in particular intragroup debt. III. The evolution of indebtedness is closely linked to investment recovery and the change in cash management Investment credit has strongly increased since 21: after a slight slowdown over the period , it took over gradually. In addition, cash credit rose sharply since the beginning of 215. A strong correlation can be observed between the dynamic of indebtedness and investment, as well as between the evolution of stocks and that of the outstanding amounts of cash credit. The analysis by category of credit should however be nuanced so as to take into account the substitution of bank credit for market financing. Figure 12 - Quarterly flows of NFCs investments, stocks and bank credit - smoothed over 4 quarters (in billions of euros) Net investment Investment credit, right scale Stock variation Credit facilities, right scale Source: Banque de France. Last observation: 217 Q2. Note: net investment corresponds to gross fixed capital formation less fixed capital consumption. 1

11 Investment and acquisitions The investment ratio of French NFCs, measured as investment over value added, has increased sharply since 212 and stands at 22% in 216, a level that represents a historical peak for the last 3 years. However, the self-financing ratio (i.e. the ratio of savings over investment flows and changes in stocks) has considerably decreased since the end of 199 s when it reached a peak of 1%. The corporate savings cover nowadays only up to 8% of the financing of investment and stock changes, which explains the recourse to debt. An important share of large enterprises debt is used to finance acquisitions, especially abroad. In 216, the debt of large enterprises increased by 87.2 billion ( 77 billion in 215) and the need of financing linked to French direct investment abroad increased by 18 billion, to a level of 52.1 billion ( 34 billion in 215). In addition, some of the acquisitions take place through leverage buy-out (LBO) financing and relate to important volumes of debt (see the Box). Box Risks related to NFCs leverage buy-out (LBO) financing The recent period has been characterized by a strong dynamism of structured issues and LBO transactions, although still at lower levels than in 28. Like at that time, LBO transactions were driven by historically low interest rates, abundant liquidity and strong competition explained by the high profitability of these activities. By contrast, the current transactions present some specificities: (i) a gradual loosening of covenants; (ii) a larger importance of operations of renewal of credit already extended (debt rollover), sometimes under more aggressive conditions; (iii) an increase in the share of non-bank financing. The default rate of LBO financed by French banks is rather low and remains under the average of the SSM (around 8% compared to 1.2%), but is sensitive to a reversal of the market situation, especially given that a large part of loans reach maturity in Cash management The increase in NFC s debt has to be assessed in the context of a large increase in cash holdings ( 579 billion in 217 Q2) that more than doubled since 28 Q1 (+ 33 billion). A significant increase took place in deposits and cash held by NFCs as a share of their gross operating surplus, from 3% in 28 to more than 625% in 217 Q2. This way, net indebtedness of firms (i.e. debt minus cash holdings), as measured by the Banque de France, registered a limited increase since 27 (48.6% of GDP in 217 Q2 compared to 45.9% in 28). However, the average ratio of net indebtedness of the 8 largest listed French groups has increased in 216, due to the rise in gross financial indebtedness (+9.2%). The increase in financial indebtedness remains however under control as it is accompanied by an increase in cash holdings (+3.5%) that contributes to a rise in net debt and own funds (+ 4.4%). However, the increase in net debt, i.e. gross debt minus cash holdings, is mainly due to the groups presenting a high leverage (last quartile). Thus, even within a population of similar firms, the situations and behaviors are highly heterogeneous. The evolution of debt corresponds more to a change in cash management practices. It can be interpreted as a consequence of the development of market financing (less flexible by nature) and, more generally, as an accumulation of precautionary liquidity in the aftermath of the crisis. There is a transition of firms towards higher buffers of cash holdings. This phenomenon is reinforced on one side by changes in the regulatory and institutional frameworks supporting the development of corporate bond markets (in particular the development of PP euro and of bond funds, the reform of the legal framework) and, on the other side, by the centralization of cash management of large groups in France. These developments take place in a context of a significant drop in interest rates that was coupled with a decrease in spreads and a decline in carry costs. With the extension of the maturity (to 2 years or 11

12 more), firms face today a lower cost of financing for reinforcing their cash holdings, and can this way react in case of investment opportunities. The decrease in carry costs could even feed opportunistic behaviors (carry trade) if the return on deposits and liquidity is higher than the borrowing cost. Some elements, in particular the rise in the volume of issuances of BT corporates at negative rates for all maturities from 1 day to one year raise some questions on the development of such a strategy. IV. Leverage developments are under control, except for large enterprises, and the goodwill still plays a significant role The leverage of firms (debt over own funds) is not declining on average, reflecting a sharp increase in firms equity along with the rise of debt. As a matter of fact, an upward revision of firms margins reinforced the amount of equity. In addition, the accumulation of liquidity triggered a decrease in net leverage. We should however underline the existence of rather heterogeneous evolutions in leverage, in particular depending on firms size. According to data aggregated over the sample, the ratios of global net and gross leverage (i.e. the sum of debt over the sum of equity 7 ) have considerably declined since 212, to reach 49% and 82% respectively in 216 (6% and 84% respectively in 212, figure 15). On the opposite, for large enterprises, net leverage has declined by 9 basis points (from 64% to 55%) while gross leverage increased by 3 basis points (from 83% to 86%). Figure 13 Ratios of weighted gross leverage, NFCs (by firm size) SME 3-Large enterprises 2-Intermediate sized enterprises AA-Total Note: NFCs debt is consolidated at the level of the institutional sector, 3rd approach (see Appendix). Source: Banque de France. Last observation: 216. The increase in NFCs debt corresponds in particular to external growth operations that lead to the registration of goodwill 8 in the assets of the acquiring company. For some firms, the level of equity remains sensitive to the valuation of the goodwill. 7 This corresponds to the average of leverage ratios weighted by the relative share of each group in total own funds of the entire sample. 8 The goodwill corresponds to the difference between the purchase price of the target and the accounting value of its own funds. 12

13 Figure 14 The evolution of goodwill and financing sources of the 2 most indebted groups (basis 1 in 212) Shareholders' equity Net debt Goodwill Note: data on the 2 most indebted groups out of a panel of 33 groups that report IFRS consolidated data. Source: Banque de France. Last observation: 216. Since 213, the recovery of the goodwill relative to total equity remains limited, the average of goodwillto-equity ratio stands in 216 below the level reached in 212 (55% in 216, 6% in 212). However, an important share of firms with a high goodwill-to-equity ratio belongs to the most indebted firms, consistent with the fact that debt is mostly associated with external growth operations, while investment is rather self-financed. This way, the most costly acquisitions are associated to larger indebtedness. Figure 15 The distribution of the ratio of goodwill to total equity (in percentage) Third quartile Median First quartile Mean Note: data on the 2 most indebted groups out of a panel of 33 groups that report IFRS consolidated data. Source: Banque de France. Last observation: 216. Appendix The measures of NFCs indebtedness Several measures of NFCs indebtedness exist. They differ depending on the level of consolidation, i.e. the way firms balance sheet is retreated so as to withdraw some loans extended among NFCs and that could not be appropriate for the macroeconomic analysis of firms indebtedness. The unconsolidated debt of NFCs (published by Insee and Eurostat, of 128.5% of GDP as of end- 216) represents the broad approach. It is the sum of all debt in the liabilities of NFCs: credit (F4), 13

14 debt securities (F3). It presents the advantage of a simple definition of NFCs debt that does not include the category other payable accounts (F8), in particular supplier debt. By contrast, it includes all debt extended between two NFCs, which is rather problematic when they result solely from an accounting entry. This is the case when a parent firm finances a subsidiary via a loan: the resulting debt illustrates the organizational choice of the group in subsidiaries that are legally independent (in case the parent company decides to absorb its subsidiaries, these loans would be neutralized). If the parent company finances through bank credit or the issuance of debt securities in order to finance its subsidiaries, there could be an issue of double counting. NFCs debt consolidated at the level of the institutional sector represents a narrow measure of NFCs indebtedness that aims to exclude all credit extended among NFCs so as to avoid the problem of double counting. Three approaches exist as regards credit (F4) : - The first approach (INSEE consolidation, equally used by the Eurostat: 9.1% of GDP as of end ) consists in deducting from total credit in the liabilities of NFCs all credit in the assets of NFCs that represent credit extended to other domestic NFCs. This measures requires having all the information on credit counterparties in firms balance sheets ( from whom to whom data); - The second approach (the Banque de France consolidation applied for foreign countries in situations of international comparisons) consists in deducting from total credit in the liabilities of NFCs all credit in the assets of NFCs, to all sectors, including the rest of the world. This measure does not require to detain from whom to whom data. This concept is the one retained in the current note for the international comparisons of debt levels. - The third approach (the Banque de France consolidation end-216: 71.3% of GDP) considers only credit in the assets of resident and non-resident financial institutions. This allows excluding credit extended among NFCs. According to Banque de France the results should be similar when applying the second and third method of consolidation, provided that credit extended by domestic NFCs to other NFCs from the rest of the world, subsidiaries or not (that are abusively removed in the second approach) are equal to credit extended by NFCs from the rest of the world to domestic NFCs (that are excluded from the third approach of consolidation, but not from the second one). This approach is retained in the current note for the figures related to France. As regards debt securities (F3), no consolidation is carried out. The exchangeable nature of debt securities on the secondary market does not allow assessing in a precise and accurate way which institutional sector detains NFC debt. Debt securities are expressed in nominal values. NFCs debt consolidated at the level of the group allows an «ideal» consolidation of NFCs debt that would consist in all credit extended to firms without any double counting issue. This approach is based on firms accounting and would result in a balance sheet consolidation at the group level. FIBEN database at the Banque de France allows this consolidation, including for international groups, despite the presence of some methodological approximations. Its perimeter is limited to firms with a turnover superior to 75 or below a threshold in terms of share capital (Public Limited Company, SA in French, with a capital superior to 76, Limited Liability Companies,SARL in French, with a capital superior to 35). FIBEN data, published on an annual basis, provide two indicators: The bank and market indebtedness of NFCs ( 55.7% of GDP as of end-215), which is a narrow measure that does not require consolidation at the group level; The gross financial indebtedness of NFCs (15.8% end-215), defined as the sum of debt in the liabilities that is consolidated at the group level. For each firm, the financial debt computed this way cannot be inferior to bank and market indebtedness. 9 The figures cover both F3 and F4 items. However, only the F4 is consolidated. 14

15 15