Financial integration is increasingly being seen as a tool to spur investment based growth via a reduction of the cost of equity, bond, and bank finance. Spurn by the introduction of a common currency, the euro area has been a prime example of financial integration, with various estimates putting the resulting increase in GDP resulting from it at between 0.3% and 2% (see, for example, VQuantification of the Macro&Economic Impact of EU Financial IntegrationV by London Economics). Of it, around one third has been attributed to the reduction of bank finance in the aggregate. However, while recent empirical studies have provided us with important insights into the determinants of international financial integration (Lane and Milessi& Ferretti [2008]), as well as its effects on entrepreneurial activity (Ongena and Gianetti [2005] and Alfaro and Charlton [2007]) and economic growth (Edison et al. [2002]), little is known of the micro effects of financial integration, notably of its effects on bank loans to individual firms. Our paper aims at filling this gap by using monthly data on integration and firm level loan characteristics in 8 new members of the EU to study the effect of interbank money markets integration on the cost, maturity, collateral requirements and currency denomination of small business loans.
How is interbank money markets integration expected to affect the bank loans conditions? Theory implies that this type of financial integration reduces the cost and increases the availability of external funding to business firms via three different channels: 1) increased competition for supplying loans in the banking sector, 2) reduced cost of external funding to banks, and 3) reduced transaction costs. As the interbank market is the most convenient way for banks to acquire short& and long&term loans to finance their own investment operations, a more integrated money market implies that banks can supply more funds at the same cost, or the same volume of loans at a lower cost. Hence, more favourable conditions to the bank as a borrower will be expected to result in more favourable conditions business firms which borrow from the bank, notably in terms of the rates at which funds are lent.
Moreover, apart from adjusting the direct cost of the loan, banks can react to a change in the money market rates by adjusting the cost of the loans along other dimensions, such as loan maturity and collateral requirements, and this effect has been observed empirically too (Heineman and Schuler [2002]). In particular, banks may be more willing to issue loans with a very short or very long maturity if their own borrowing conditions have become more favourable. Also, collateral requirements can be relaxed in response to cheaper finance for banks, but they could also be raised if the distribution of a larger volume of funds leads to the financing of more marginally profitable projects (Inderst and Mueller [2006]). It has also been argued that stricter collateral requirements may redress the incorrect emphasis of cheap credit over project screening when market imperfections are present in the banking industry (Manove et al. [2001]). In terms of bank er ciency, the time necessary to negotiate a loan may go down as a result of increased competition on the supply side, but it could also go up if screening has increased in response to the higher volume of funds distributed. Finally, firms may choose more often loans denominated in local currency if as a result from money markets integration the cost of those has decreased relative to the cost of foreign currency funds (Brown et al. [2008]).
It is important to note that we will be addressing the issue of money&market integration separate from the issue of cross&border bank ownership. While the two processes may coincide in direction and magnitude over time, the idea of pure integration is that banks may use borrowing and lending in an ever more integrated money market to reduce the income shocks of individual business borrowers. By thus sharing the risk with other banks and financial institutions, banks are able to offer more favourable and stable lending conditions, and hence money market integration may have a positive effect in that direction net of foreign bank ownership. For example, along similar lines of argument, Demyanyk et al. (2007) find that the deregulation of the US banking sector affected income insurance of small business owners even without explicit multistate cross&ownership between banks.
We however also address the issue of financial integration in the context of bank foreign entry and bank competition. Regarding the former, evidence from the global market suggests that the effects of foreign bank entry are ambiguous. Foreign banks tend to increase the quality and availability of financial services in the domestic market and the countryos access to foreign capital. In addition to that, they also improve the overall quality of the loan portfolio of domestic banks because they are less susceptible to government pressure to lend to Vpreferred borrowersV, as may be the case with some domestic institutions (see, for example, Levine [1996] and Caprio and Honohan [1999]). At the same time, foreign banks may ration credit to firms, especially to small ones and those that operate in the sector for non&tradeables to a larger extent than domestic banks, either due to better ability to compete for the most creditworthy borrowers, or because of lack of soft information whose availability enables domestic banks to monitor loans to opaque firms more effectively. We test our findings with attention to the ownership structure of the domestic banking market, and find that most of the observed effects of money market integration have been realized in markets with high degree of bank asset ownership by foreign banks. We hypothesize that this may be because in the post&1989 period, foreign banks didnot enter the central and eastern European credit market by opening new local subsidiaries, but most often via the purchase of domestic banks. Hence, while the improvements in er ciency brought about by competition and higher access to foreign capital due to foreign bank presence are probably substantial, any potential loss of soft information has probably been limited.
With regards to the second issue, it is relevant as, for instance, in studying the effect of increased integration on the cost of funds to business firms, we are essentially estimating the degree of pass&through. It is commonly believed that the benefits of credit markets competition arise from lower surplus extracted from borrowers, as well as from a higher speed of interest rate adjustment (Herrmann and Jochen [2003]). With a high degree of concentration, the few banks in business will be slow to pass declining market rates to the consumers. With high competition, banks are forced to react fast or risk losing market share, and hence in such markets money market integration will affect the conditions of bank loans more rapidly and substantially. At the same time, Petersen and Rajan (1995) have argued that creditors are more likely to finance credit&constrained firms when credit markets are concentrated because it is easier for those creditors to internalize the benefits of assisting the firms. They also provided evidence to the fact that international financial liberalization has led to deterioration in the credit conditions for small firms, and argued that this is because small credit&constrained firms thrive on local banking monopolies. The evidence we present is mixed, in the sense that the gains to small and medium firms in terms of lower cost of credit, and more widely available short&term loans have been most pronounced in markets with high degrees of banking competition, consistent with the first argument, while improvements in bank er ciency (time necessary to negotiate a loan) are only observed in concentrated banking markets, implying some justification of the local banking monopoly hypothesis of Petersen and Rajan.
Finally, we investigate for a euro effect. Two of the countries in the sample (Estonia and Lithuania) had their currencies pegged to the euro over the period of interest, the central bank in a third one (Latvia) had monetary policy identical to the ECBos, and a fourth one (Slovenia) was getting ready to enter the euro zone, which it did on January 1 2007. We find a weak euro effect in the form of a higher probability of small firmsosecuring a non&collateralized loan in countries whose currency was pegged to the euro, but we find no effect on the cost of loans or, surprisingly, on the loanos currency denomination.
We find that interbank markets integration tends to increase the probability that the loan will not be collateralized for large firms, the probability that the loan will have a maturity longer than 48 months (for medium firms only), and that integration increases the probability that small firms will negotiate loans in foreign currency denomination. We also find effects that are differentiated by banking sector characteristics: money markets integration reduces the annualized cost of loans charged to small and medium enterprises in countries with high share of foreign banks by between 101 and 117 basis points, increases the probability in those of negotiating a short&term loan for both small and medium firms by about 15.1%., and decreases the time necessary to negotiate the loan to small firms by 7.7 days. We find identical results in markets characterized by low degree of banking concentration. Finally, we find that the probability of securing a non&collateralized loan increases with financial integration for countries with currencies pegged to the euro, but this result is not confirmed by all robustness tests.
The paper proceeds as follows. In Section 2 we lay out the empirical strategy. In Section 3 we describe our data on money markets integration, and in Section 4 we describe the firm level data on actual bank loans and general firm characteristics. Section 5 summarizes the empirical evidence, and Section 6 concludes.
Download
PDF Ebook The Effect of Money Markets Integration on Bank Loans
