Measuring the access to finance of small and medium-sized enterprises across the euro area through a flexible survey

Transcription

1 Measuring the access to finance of small and medium-sized enterprises across the euro area through a flexible survey M. Osiewicz and S. Pérez-Duarte, ECB ECB workshop, 6 December 211

2 Initial disclaimer

3 The issues SMEs and large firms respond differently to monetary policy (more bank-dependent, little access to bond or equity markets) SMEs are a significant fraction of the euro area (99.8% in terms of firms, 67% in terms of employment) Little to no timely statistics on access to finance for SMEs, certainly not comparable across the euro area TANKAN (Japan), PMI (US), ECFIN business surveys (EU), country specific surveys

4 Examples outside the euro area US: National Federation of Independent Business (NFIB), Small Business Economic Trends Japan: TANKAN, Short-term Economic Survey of Enterprises UK SME Business Barometer

5 The solution: sample and questionnaire The survey on the access to finance of SMEs is interest of ECB and European Commission, but slightly different objectives euro area (6 months) vs. (large) European Union (2 years) short term vs. more structural analysis flexible geometry: countries, questions, sample size euro area sampling strategy: smallest euro area countries not included. Change in smaller countries in March 211 questionnaire improvement: better coverage of bank overdrafts and credit cards; clarifying not used and not relevant (try to be) comparable with BLS and ECFIN business surveys: qualitative focus

6 Sample design 1 interviews in each of the four largest countries of the euro area 35 interviews in the other euro area countries (every two years: 43) EE, CY, LU, MT, SI, SK represent less than 3% of euro area and are included only every two years 1% large firms; 3% each of micro, small and medium Oversampling of small, medium and large in terms of number of firms, undersampling of large firms in terms of turnover Use of sampling weights to correct this

7 Sample design (continued) Current approach strategy inherited from Eurobarometer Sample drawn from Dun & Bradstreet (somewhat comparable, OK quality in the euro area) Stratified by country and firm size, post-stratified by sector Weights proxy for economic importance using employment (at cell = country x sector x size level) How to model this quota sampling in statistical terms?

8 The solution: production and analysis On the production side: Timeliness crucial! Telephone one month of fieldwork, 7,5 interviews Outsourced (first two waves: Gallup, following ones: IPSOS) Streamlined data processing integrated validation and production, controlled by one metadata database On the analysis side: Quick delivery of flat report Survey allows total flexibility for microanalysis, feeds into other notes and reports

9 Please don t hang up! Telephone only mixed mode with Internet/fax not successful Small firms are the hardest to reach There is never a good time to call Getting through the switchboard Incentives not incentive enough Interviewus interruptus Different techniques to be tested

10 Low response rate for first contacts 1 Response Rate 4 Refusal Rate 2 Contact Rate BE DE IT AT ES SI FR PT GR EE IE MT FI Source: EC-ECB survey

11 Higher response rate of panel firms 1 Response Rate 4 Refusal Rate 2 Contact Rate GR PT AT DE IT BE ES NL FI IE FR Source: EC-ECB survey

12 Panel retention (wave 5) 7 Many firms accept being re-interviewed Austria Belgium Finland France Germany Greece Ireland Italy Netherlands Portugal Spain Average but not all are successfully recontacted Non-panellist Panellist Total Source: EC-ECB survey, wave 211 Austria Belgium Finland France Germany Greece Ireland Italy Netherlands Portugal Spain

13 The complicated panel structure of the SAFE 8, 7, 6, 5, 4, 3, 2, 1, Unique 3,4 2,4 2,3 2,3,4 1,2 1,2,4 1,2,3 1st wave 2nd wave 3rd wave 4th wave 1,2,3,4 Source: EC-ECB survey

14 If you want to use the SAFE Use the weights! Think about how you should treat Don t know and Not applicable to my firm Be aware of filtered questions and the different possibilities opinion on availability of external finance is different if the firm has actively sought it Beware of small sample sizes in some countries, especially filtered questions and/or some sector/size combinations Try to correct for design effects (approximately 1.4 for unfiltered questions)

15 Confidence intervals 5% 4% 3% 2% 1% % -1% -2% -3% -4% -5% a b c d e f a b c d e a b c d a b c d e f g h a b c d e f a b c d e a b c d e f Q2 Q3 Q5 Q6 Q11 Q23 Q9 Q1

16 Confidence intervals, by country Net percentage, availability of bank loans, SMEs 2% % -2% -4% -6% -8% DE ES FR IT AT PT IE BE GR NL FI Source: EC-ECB survey, 21 wave

17 Dissemination of results Report published by the ECB + report by the European Commission every two years Complete aggregate tables for all variables and various breakdowns available on the ECB website All time series available in the SDW Anonymised microdata available to external researchers ( only )

18 Other uses of the survey Move away from standard boundaries Composite indicators Needs, availability, or combined (tightening) Matching with corporate balance sheet Project ongoing

19 Size of firm and access to finance Sources of external financing Bank overdraft Bank loan Trade credit Leasing, factoring or more Source: EC/ECB survey on the access to finance of euro area SMEs, 211 wave

20 Many thanks for your attention Contact us for any question

21 Some results from the latest wave of the SAFE Economic sentiment Needs and availability of external financing Application to external financing and success Expectations EC questions in this wave

22 Changes in economic conditions H2 H2 H2 H2 H2 H2 H2 H2 H2 H2 H2 H2 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 Turnover Labour costs Other costs Net interest Profit Debt-to- Source: EC-ECB survey

23 External financing needs and availability Net percentages H2 H2 H2 H Bank loans Trade credit increased unchanged decreased not applicable don't know net Net percentages H2 H2 H2 H Bank loans Trade credit improved unchanged deteriorated don't know not applicable

24 Change in bank loan needs and availability Needs '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 BE DE IE GR ES FR IT NL AT PT FI 2 Availability '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 '9 '1 '11 BE DE IE GR ES FR IT NL AT PT FI

25 Application to external finance Stable success rate in getting loans H2 H2 H2 H Bank loan (new or renewal) Trade credit don't know did not apply for other reasons did not apply because of sufficient internal funds did not apply because of possible rejection applied H2 H2 H2 H Bank loan (new or renewal) Trade credit don't know application rejected application granted but cost too high application granted in part application granted in full

26 hides large country differences BE DE IE GR ES FR IT NL AT PT FI Total Applied and got everything Applied and got most of it Applied and got a limited part of it Applied but refused because cost too high Applied but was rejected

27 Expectations of availability Net percentages H2 H2 H2 H2 H2 H Internal funds Bank loans Trade credit don't know not applicable expected to deteriorate expected to remain unchanged expected to improve

28 Most limiting factor Most limiting factor to get bank or other loans BE DE IE GR ES FR IT NL AT PT FI Other small Euro area euro area countries Don't know There are no obstacles Interest rates or price too high Insufficient collateral or guarantee Financing not available at all Other

29 Many thanks for your attention