Interesting insight from Tabitha Wilde, Yale School of Management. Small and young businesses depend on external financing for their growth and survival. It is believed that local or decentralized banks always offer better lending terms to these firms, however a new study from Rodrigo Canales of the Yale School of Management and Ramana Nanda of Harvard Business School finds that this is only true when banks face local competition. Small banks are thought to be best for small businesses because their decentralized lending structure gives branch managers autonomy over lending decisions, which gives them flexibility to adapt to local business needs. In contrast, large centralized banks make decisions based on credit scoring models that cannot account for specific small business characteristics, such as their common limitations in providing formal financial reports. “We show that it is true that decentralized banks create better lending outcomes for small firms than centralized banks, precisely because of the discretion granted to local branch managers,” said Canales. “The huge caveat is that this is only true when there’s competition.” Without competition, the same discretion that allows branch managers to shape lending decisions to the local environment also allows them to exploit their market power by cherry-picking the best firms, giving them smaller loans, and charging higher interest rates. Canales and Nanda analyzed all of the loans that were given to small firms in Mexico from 2002 to 2006. All of the lenders in the data set are large, national banks. Some follow a centralized model, while others follow a decentralized model. Focusing on large banks that use different models allowed the authors to prove that it is the structure of banks, rather than their size, that drives lending outcomes. The data show that in markets with less competition, decentralized banks gave loans that were 60% smaller than those given by centralized banks; they lent to larger, safer firms; and they charged higher interest rates. “Branch managers will be reactive to their local environment and that can be good or bad for firms,” said Canales. “If the decentralized bank is the only game in town, the manager will give firms worse conditions because he or she knows the firms don’t have other options. If there is competition, the manager can react to that and operate in ways that centralized competitors cannot.” The results help explain, for example, why entrepreneurship increased following the U.S. banking deregulation that led to mergers and acquisitions that decreased the number of small banks. “Before banking deregulation, all banks were regional, and all banks were decentralized,” said Canales. “You would expect that lending to small firms would have dropped when banks were able to become centralized and nationalized. The increase in lending to small businesses may have occurred because small banks no longer had local monopolies, and the addition of market competition benefited entrepreneurs.” The study has implications for how to help small businesses. “If you want to help small firms, then you want to foster competition, and you want to foster a variety of lending models instead of just focusing on centralized or decentralized lending for small business,” said Canales. The paper, “A Darker Side to Decentralized Banks: Market Power and Credit Rationing in SME Lending,” is published in the Journal of Financial Economics (August 2012).