adobestock189748797jpeg
VoxEU Column COVID-19 EU policies Labour Markets

The impact of public sector lending to SMEs on employment and investment

With the coronavirus crisis unfolding, many countries have announced new lending and guarantee programmes dedicated to supporting businesses’ access to finance. This column examines the impact of such programmes, focusing on European Investment Bank lending schemes. The findings suggest that publicly funded lending support programmes can make a difference in maintaining employment and investment activity at the firm level.

With the coronavirus crisis unfolding, many countries in the EU and elsewhere have been announcing new lending and guarantee programmes, dedicated to supporting businesses’ access to finance, as part of their efforts to mitigate the negative impact of the pandemic on economic activity and employment (OECD 2020). Beyond national initiatives, international financial institutions (IFIs) such as the EBRD, the European Investment Bank (EIB) and the International Finance Corporation (IFC) have also launched new instruments dedicated to support lending to small and medium-sized enterprises (SMEs) and mid-caps during this difficult period.

It is therefore relevant from a policy perspective to ask whether policymakers can expect these tools to have the intended impact. More precisely, the question is whether such public support to SME lending – funded either by national promotional institutions or by international financial institutions – has a measurable influence on the beneficiary firms’ performance. In particular, we ask whether these companies have been faring better in terms of employment, profitability, growth or investment than other similar firms that have not benefitted from such support.

Assessing the impact of public lending support to SMEs

Due to practical considerations, such as scale and implementation complexity, more often than not such programmes cannot be evaluated in controlled environments. As a consequence, the true causal mechanisms may be blurred in the data by a common-factor or selection bias, affecting both the support-granting and support-receiving entities. While the identification of true causal linkages remains a challenge, the emerging literature on the impact of public support to SME lending attempts to overcome the measurement problems, and the bulk of these studies call for optimism. 

For instance, Brown and Earle (2017) analyse the employment performance of 157,400 firms that benefited from loan guarantee programmes of the US Small Business Administration (SBA) between 1990 and 2009. The counterfactual sample is constructed using a combination of propensity score matching and difference-in differences.  The results suggest an increase of between 3 and 3.5 jobs for each million dollars of loans in the first three years after loan disbursement, compared to similar firms that have not received SBA-supported financing. In the European context, Brault and Signore (2019) find that guaranteed loans under the EU Multi-Annual Programme and the Competitiveness and Innovation Framework Programme between 2002 to 2016 positively affected growth of firms’ assets, share of intangible assets, sales and employment, and lowered the probability of default metrics. Our paper with Raschid Amamou (Amamou et al. 2020) contributes to the literature by assessing the impact of funding from the EIB to SMEs on a large and multi-country set of entities covering 27 out of 28 EU countries.

Intermediated lending by the EIB

The main question we put under the microscope is to what extent the firms that participated in EIB-supported lending schemes across the EU member states performed differently in terms of employment and investment activity, should they not have had such opportunities. We build on earlier work by Gereben et al. (2019), who carried out a similar analysis on a smaller data set, focusing on countries of Central and Eastern Europe (CEE).

EIB funding products targeting SMEs take the form of Multiple Beneficiary Intermediated Loans (MBILs), using an intermediated lending model involving private banks and leasing companies. The EIB provides funding to these intermediaries at better-than-market conditions, while they commit to use the funds to extend loans to SMEs, and to partially transfer the financial advantage to the final beneficiaries in the form of an interest rate reduction and/or the provision of longer tenors.

The EIB’s support through intermediated lending may translate to better SME performance through two distinct channels. First, financial intermediaries pass some of the received funding advantage to borrowing firms in the form of lower financing cost or longer maturity. This extra advantage can contribute to better economic performance due to strengthened profit and loss account. Second, public sector support might alleviate credit constraints on the banks’ funding side. Particularly in times of economic or financial downturns, when capital, liquidity or both are scarce, a public line of credit expands the funding base of banks, making it possible to lend to firms with viable investment projects, which would otherwise have been rejected, or only partially served, due to lack of funding.

While the lending agreement requires the banks and financial institutions to report on the final beneficiary firms, the records do not contain data on their financial or economic performance. Therefore, we merge our data, covering EIB-supported loan allocations between 2008 and 2015, with Bureau van Dijk’s Orbis/Amadeus data set. Overall, we work with a sample of about 65,000 firms, taking into account their financial and economic performance in the years before and after the disbursement of the loans.1

To create a counterfactual sample, we pair each EIB beneficiary with a company without an EIB tag, but having otherwise similar financial and non-financial records, using propensity score matching (PSM). In the next step, we use difference-in-differences (DID) to evaluate performance of the beneficiaries relative to the counterfactual sample.  The combination of PSM and DID allows us to control for potential confounders, therefore the performance difference between the two groups of firms can be given causal interpretation.

Our results suggest a positive impact on employment and investment

We find a significant positive impact on employment in the three years that follow the allocation of the loan. The positive results on employment suggest that access to external funding at advantageous conditions improves the economic situation of receiving firms to the extent where they are more likely to keep employees and/or hire new ones than firms without EIB support. The size of the coefficient indicates that firms receiving EIB lending increased, on average, their employment 4–6% more relative to the peer group of firms without EIB financing (Figure 1). We also find a positive impact on fixed assets, in the range of 8% to 14%, indicating that the beneficiaries typically used the loans disbursed to purchase investment goods. Our analysis indicates that the impact of EIB lending on employment and fixed assets was somewhat higher in the CEE and South Europe countries than in the West and North Europe.

Figure 1 Estimates of the EIB loan’s effect year by year

Overall, our results support the view that publicly funded lending support programmes can make a difference in maintaining employment and investment activity at the firm level. The results matter for economic policy design in the context of the current economic downturn, as they confirm that EIB-intermediated lending schemes have been successful in terms of influencing the economic decisions of the final beneficiaries, and suggest that the resulting know-how may prove effective when designing instruments aiming at mitigating the impact of the COVID pandemic.

There is an important caveat that we need to mention. Our results, in line with the literature, provide partial equilibrium analysis. They reveal the direct impact, but do not consider redistribution effects, costs, and overall impact on welfare, such as the fiscal costs of such guarantees. In the current context, while guarantees may help with supporting employment, they will inevitably be costly for taxpayers due to the potentially high level of corporate defaults. While general equilibrium effects and welfare analysis go beyond the scope of our analysis, we would like to point the interested reader to the paper by Anginer et al. (2011), who discuss welfare implications of public credit guarantees. The authors argue that such guarantee schemes can be welfare enhancing in the presence of aggregate and non-diversifiable risk, by helping to resolve collective action failures. In this respect, public support schemes are similar to deposit guarantees and lender-of-last-resort facilities, which aim to avoid the magnification of systemic events.

Authors’ note:  The views expressed in this column are those of the authors and do not necessarily reflect the position of the European Investment Bank. The authors would like to thank Debora Revoltella and Tim Bending for their useful comments, and Alessandro Barbera for his research assistance.

References

Amamou, R, Á Gereben and M Wolski (2020), “Making a difference: Assessing the impact of the EIB's funding to SMEs”, EIB Working Paper.

Anginer, D, A de la Torre and A Ize (2011), “Risk Absorption by the State: When is it Good Public Policy?”, World Bank Policy Research Working Paper no. 5893.

Brault, J and S Signore (2019), "The real effects of EU loan guarantee schemes for SMEs: A pan-European assessment", EIF Working Paper Series 2019/56.

Brown, J D and J S Earle (2017), “Finance and Growth at the Firm Level: Evidence from SBA Loans”, The Journal of Finance 72: 1039-1080.

Gereben, A, A Rop, M Petricek and A Winkler (2019), “Do IFIs make a difference? The impact of EIB lending support for SMEs in Central and Eastern Europe during the global financial crisis”, EIB Working Paper 2019/09.

OECD (2020), “Covid-19 : SME Policy Responses”, OECD Centre for Entrepreneurship, SMEs, Regions and Cities.

Endnote

1 Since the publication of the paper (Amamou et al. 2020), we were able to add allocations from 2015 to the analysis. Adding the most recent data has not affected our key findings. We report the quantitative results here based on this extended data set.

1,260 Reads