【正文】 on average, the growth of these firms is onetoone related to retained profits. Secondly, while finance constraints seem to be less binding for mediumsized enterprises, their growth, in parison to the growth of large firms, nevertheless depends more on theavailability of internal funds. Thirdly, highly leveraged firms have greater difficulties in tapping external finance and, hence, exploiting their growth potential. How could one possibly improve the supply of finance to SMEs? It is useful to distinguish between public policy measures and fforts that lenders and borrowers can make to alleviate finance constraints. Wagenvoort briefly reviews the literature on the effectiveness of public lending programmes and guarantee main conclusion is that while direct lending and guarantee programmes usually benefit the recipients and help ease finance constraints, it has been questioned whether they improve the allocation of resources in an economy. Nevertheless a positive net return on public intervention can be expected if intervention reduces information asymmetries between borrowers and lenders and thus helps solving information problems. For instance, public authorities may stimulate information sharing among lenders. A recent study (Jappelli and Pagano 2002) shows that information sharing among lenders increases bank lending and reduces credit risk. Borrowers and lenders themselves can also contribute to solve finance problems of SMEs by reducing information asymmetries directly. As argued above, the establishment of longterm relationships has the potential to achieve this. 4. Relationship banking and bank consolidationIs there empirical evidence to support the view that relationship banking can mitigate finance constraints? Ongena and Smith (2000) report substantial variation in the average number of bankfirm relationships across European countries. The three country studies reviewed here confirm this result and they show that firms make considerable use of multiple banking. Guiso’s analysis reveals that in Italy small firms keep on average more than four bank relationships whereas large Italian firms diversify their credit needs over more than 10 credit institutions. As shown by Hommel and Schneider, the Mittelstand in Germany relies on a smaller number of bank ties but even the small German firms on average borrow from more than one lender. Very small German firms borrow on average from two banks whereas large Why is it then that SMEs keep fewer and shorter bank relationships than large firms? As credit availability improves when relationships bee longer,one would expect information opaque SMEs to stay with the same creditor(s). To begin with the number of relationships, as Dietsch notes, an obvious reason is that SMEs have to spread out fixed costs of lending over a smaller loan amount. Adding more creditors to the list of the firm’s financial intermediaries will trigger additional costs. Therefore, smaller firms may be less willing to borrow from several banks at the same time. However, the disadvantage of relying only on one bank is that this bank may turn into a monopolist over time. Dietsch explains that, although it is expensive for the smaller firms to provoke petitive behaviour of their lenders by maintaining multiple relationships, smaller firms may still break monopolies by switching banks when time passes. This may explain the relatively short duration of bankfirm relationships of smaller firms. One remark is called for. Hommel and Schneider point out that the number of initial credit offers a firm enquires about before finalising a loan contract may be more informative than the number of its relationships. This is especially the case if firms seek offers from banks they had no prior relationship with. Another important element is whether firms seek offers from banks that are not located in the area where the firms have their headquarters. Overall, the authors conclude that Mittelstand firms seem to be more flexible than monly assumed. Companies approaching several banks obtain an average of approximately three loan offers. What is more, a substantial amount of offers originates from banks that had no prior relationship with the firm and/or from banks situated outside the immediate geographical vicinity of the firm seeking finance. This is quite surprising because it is often argued that a local bank is best informed about firms in its region, essentially tying small firms to local banks. Having established that both relationship banking and multiple banking enhance credit availability for SMEs, Dietsch continues his analysis by investigating whether bank consolidation in France has altered those two important features of European banking. He emphasises that bank consolidation in France went hand in hand with a lower concentration level in the business loan market. The wave of mergers and acquisitions thus seems to have stimulated petition between credit shows that the number o