Firms’ Financing Constraints - Do Perceptions Match the Actual Situation?
By Annalisa Ferrando (European Central Bank) and Klaas Mulier (Ghent University)
The financial crisis has renewed the interest of researchers and policy makers in the link between the financial economy and the real economy. In his speech of 13 December 2012, Benoît Coeuré, Executive Board Member of the European Central Bank (ECB), said that “the financing environment and access to finance for euro area corporates are important elements in the policy-making process of the ECB, both for standard and non-standard monetary policy decisions”. The motivation for this statement is that frictions and dysfunctionalities in financial intermediation and the transmission mechanism influence firm investment, and hence growth and price stability.
In his speech, Benoît Coeuré further stated that “the ECB regards small and medium-sized enterprises (SMEs) as a crucial component of the euro area economy ... Moreover, these SMEs are highly dependent on bank loans which are supplied at a higher cost and with a higher rejection rate than for larger firms, due to asymmetric information”. The ECB board member thus highlights the importance of SMEs in the euro area. This is not surprising since more than 99% of the European firms are SMEs, but he also acknowledges that they might face different financial conditions than large firms, as they are more likely to suffer from financial market frictions. This can be seen in Figure 1 where it is shown that the spread between small loans (< 1 million €), which are assumed to be granted mainly to SMEs, and large loans (> 1 million €) is positive; and has increased and remained elevated since the onset of the financial crisis.
Figure 1: Spread between bank lending rates on small and large loans to non-financial corporations
Source: ECB Monthly Bulletin, July 2013, box 6
Notes: Spread in basis points between small loans (less than 1 million euro) and large loans (more than 1 million euro).
Therefore, this paper investigates which firm characteristics (financial and non-financial) are correlated with the firm's self-reported financing constraints -of mainly SMEs- in order to get a better understanding of the nature of financing constraints during the recent financial crisis. Is the observed reduction in lending during the crisis merely a consequence of the tensions in financial and banking markets? Is there still intermediation towards the most promising firms? Do SMEs suffer more just because they are smaller or because they are riskier?
The two main data sources for our analysis are the survey on access to finance of small and medium-sized enterprises (SAFE)1 which is conducted by the ECB and the European Commission and the AMADEUS database which is gathered by Bureau van Dijk. We use a nearest neighbour non-parametric matching procedure to match the firms in SAFE with financial accounts taken from the large AMADEUS dataset which contains balance sheet and profit & loss account information of firms. This way we obtain a unique dataset with direct information on the firm's self-reported financing constraints combined with financial and non-financial characteristics, which will allow us to investigate whether the firms that self-report to be financially constrained have different characteristics than financially unconstrained firms.
We derive two measures of financing constraints from the survey: perceived financing constraints and actual financing constraints. Perceived financing constraints are measured through firms' self-assessment on whether access to finance constitutes their most pressing problem or not. These are the firms that choose "Access to Finance" from a set of potential problems (in question q0) ranging from finding customers and the presence of competition to increased costs of production or labour and the presence of regulation. This reply may be only based on the general perception of the respondent and is not a priori based on its actual experience; therefore we call this measure the perceived financing constraints.
The actual financing constraints are a more objective measure of financing constraints which are related to the firms' actual applications for external financing. In the survey, respondents are being asked whether they have applied for external finance or not (and why?) (question q7a) and whether they were successful or not (and why?) (question q7b).
Figure 2: Perceived and actual financing constraints
Notes: Perceived financing constraints shows percentage of respondents that chose 'Access to finance' as most pressing problem in question q0 of the SAFE survey. Actual financing constraints shows percentage of respondents that was denied access to finance conditional on having requested external finance in questions q7a and q7b of the SAFE survey.
Figure 2 shows the fragmentation of euro area financing conditions. Firms in the southern jurisdictions tend to face the highest financing constraints, both in terms of their perception and in terms of their actual applications for external finance. Firms in Belgium, Germany and especially Finland face low financing constraints, while Spanish, Greek and Portuguese firms face high financing constraints. Remarkably, in all countries it appears that a smaller percentage of firms perceive access to finance problematic while comparatively a higher percentage is denied access to external finance if they actually apply for it.
Our empirical analysis aims to investigate the existence of underlying factors that determine both firms' perception of financing constraints and firms' actual financing constraints. In particular we are interested to analyse the relative importance of financial characteristics (as derived from balance sheet and profit and loss accounts) versus non-financial characteristics (as derived from the survey). For this reason we model the probability of firms facing financing constraints as a linear function of the characteristics available from our two different data sources:
FinancialRatio is a set of ratios that summarises the financial conditions of the firm: we use profitability measures (return on equity, interest coverage ratio, profit margin), liquidity measures (working capital, working capital requirement, cash) and leverage measures (debt to assets, debt minus cash to assets, maturity of the debt). FirmControls are the non-financial characteristics of the firm such as the age, the size and the ownership structure of the firm. Finally, we also control for the Country of the firm, the Sector of the firm and the Year in which the firm responded to the survey.
The results reveal that the measures related to the profitability of the firm are more significant and robust in predicting the actual financing constraints encountered by firms than liquidity or leverage ratios. Firms with higher return on equity, higher profit margins and higher coverage ratios are less likely to have their actual application for external finance rejected. The liquidity and leverage measures appear to have some explanatory power for the actual financing constraints, however, this disappears when we control for the age and size of the firms. Further, analyzing the perception of financing constraints, the regressions show that only firms that finance a higher share of their assets with short term debt are more likely to perceive access to finance as problematic. This is due to the fact that these firms need to roll over a high share of their debt yearly and they expect that this might become very difficult or costly when market conditions turn for the worse. Additionally, we find that firm age, but surprisingly not size, is negatively related with both perceived and actual access to external finance. We show that this can be due to the fact that small firms appear to select themselves out of the loan-application process
Finally, our control variables confirm some stylized facts. The ownership dummy is significant and positively related to both perceived financing problems and actual financing obstacles. This stresses the importance of the role of ownership and the existence of internal capital markets for the financial constraints that firms belonging to groups (do not) face. Country dummies with firms located in Spain and Greece are facing significantly higher constraints than Belgian firms (which is our reference country in the estimation). This shows that there is fragmentation in the euro area which implies that firms with identical financial and non-financial characteristics (as we have defined) have higher probabilities of facing financing constraints in Spain or Greece than in Belgium. Further, controlling for the characteristics of the firm, access to finance appears to be equal across sectors. As for the time dummies, it is found that firms are more likely to face actual obstacles in 2010 and 2011 relative to 2009. The time dummies show no significant differences across time concerning the perceived access to finance.
To sum up, we find that financial characteristics are able to explain self-reported financing constraints by firms. This implies that firms should thoroughly consider their financial decisions. However, also firm age plays a large role for financing constraints. Small firms appear not to apply for external finance ‘out of fear of rejection’, although we find no evidence that they have different financial characteristics. Therefore, they are not less likely to obtain finance than other firms with the same age or financial characteristics and thus they should be encouraged to actually apply.
The main aim of this paper was to investigate the role of financial and non-financial firm characteristics to get a better understanding of the nature of perceived and actual financing constraints during the recent financial crisis. Its novelty is related to the availability of a unique dataset containing direct information on financing constraints as reported by firms in the Survey on Access to Finance of small and medium-sized Enterprises (SAFE) and the financial characteristics of those firms. To obtain this dataset we use a non-parametric matching procedure to match 11886 firms from the SAFE survey dataset with their balance sheet information out of the AMADEUS dataset with 2.3 million firms.
Perceived financing constraints are measured through firms' self-assessment on whether access to finance constitutes their most pressing problem. We also consider a more objective measure of financing obstacles which is related to the firms' actual applications for external financing. It is then investigated whether the firms that self-report to be financially constrained have different characteristics than financially unconstrained firms.
Our empirical results based on a bivariate probit model show that various measures related to the profitability of the firm are more significant and robust in predicting the financing constraints encountered by firms than liquidity or leverage ratios. Further, firms that finance a higher share of their assets with short term loans are more likely to perceive access to finance as problematic. This is due to the fact that these firms need to roll over a high share of their debt yearly and they expect that this might become very difficult or costly when market conditions turn for the worse. Finally, we show that firm age, but surprisingly not size, is negatively related with perceived and actual financing constraints. We have argued that the surprising result with respect to size can be due to the fact that small firms appear to self-select them out of the loan-application process due to ‘fear of rejection’.
The results indicate that firms should thoroughly consider their financial decisions. Still, policy makers should be aware that firms may also be discriminated on the basis of age. Further research is desirable to confirm the peculiar role that size might play for the self-selection out of the loan-application process due to ‘fear of rejection’.
Reference: Ferrando, Annalisa & Mulier, Klaas, (2013), "Firms’ financing constraints: do perceptions match the actual situation?", Working Paper Series 1577, European Central Bank.
A report containing the main results of the survey is published in the ECB website every six months. More information regarding the survey as well as the reports on the individual waves