Financial frictions and the SME investment gap: new survey evidence for Ireland

ABSTRACT

In this paper, we use novel firm-level survey data on investment expenditure for Irish SMEs to investigate the existence of investment constraints between 2016 and 2018. We model investment as a function of firms’ economic fundamentals and factors affecting firms’ access to finance. We use an empirical version of the traditional fundamental Q model which links investment to the marginal returns to capital. Empirically we deploy a two-stage Heckman model and include measures of financial frictions such as access to collateral, non-bank financing, internal finance and indebtedness. We find evidence that investment is substantially lower than the level suggested by economic fundamentals alone and that a considerable financing gap is evident. On average, we find that investment could be 55% higher than present if financing constraints were completely eliminated.

KEYWORDS:

• SMEs
• financing constraints
• investment
• Ireland
• market failures

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Many SMEs do not invest and this leads to a bunching of the investment distribution at zero. This behaviour needs to be modelled more formally.

2. Given this is a standard model, we do not reproduce the calculations to get this Lagrange multiplier. These can be seen in numerous other papers such as O’Toole, Newman, and Hennessy (2014). If readers would like the calculations, the authors can however provide these on request.

3. In practice, firms may be over investing if they have soft access to capital, due to corruption and mismanagement which may in some cases lead to the observed investment being higher than the unconstrained level. See O’Toole, Morgenroth, and Ha (2016) and Dollar and Wei (2007) for discussion.

4. See Department of Finance (2019) for further details on the survey and for the questionnaire.

5. Consequently the sample only includes a small number of SMEs which have been operating for less than two years, as these companies usually are not registered yet. Therefore, the present analysis excludes very young, likely fast-growing SMEs, for which credit constraints may be a significant problem.

6. For this purpose, price indices from Eurostat were used: 1) sector specific PPIs (for the value of turnover); 2) GDP-Gross Capital Formation deflator (for the value of total assets); and 3) GDP deflator (for the value of debt and of total investment).

7. Galindo, Schiantarelli, and Weiss (2007) also use the profits to capital ratio as a proxy for the marginal value product of capital. However, due to insufficient data we are not able to produce a consistent profits series so must instead use the sales to capital ratio.

8. Note that these weighted average percentages were obtained using the CSO size weights described above. Using sample weights, the sample includes 39% micro firms, 37% small firms and 23% medium sized firms. These differences reflect the importance of re-weighting our sample.