New York and London have been the two most dynamic and innovative world financial centres over the past two decades. Frankfurt, by contrast, remains a relatively small and undeveloped banking centre, with many German banks moving more of their operations to London in recent years.

Why should this be so? And which financial systems offer the best model for the transition economies of Eastern Europe? Recent work by Arnoud Boot and Anjan Thakor addresses these issues, asking how the structure of a country’s financial sector affects incentives for financial innovation and the efficiency of the capital markets.

The researchers focus in particular on the two characteristics that typically distinguish the financial sectors in the UK and US from that in Germany: the degree to which the banking system is functionally separated between commercial and investment banks (as in the UK and US) or universal (as in Germany); and the degree to which the financial sector is fragmented (the UK and US) as opposed to concentrated (Germany).

Boot and Thakor’s model assumes that companies choose between two types of finance for new investment: borrowing either from commercial banks or direct from the capital markets, underwritten by an investment bank.

The advantage of commercial bank borrowing is the banks’ skill and experience in post-lending monitoring of managers to prevent them investing the capital in ‘bad’ projects – those which benefit them but not shareholders. This ‘moral hazard’ problem is particularly important in countries with unsophisticated financial markets and for companies where there is no publicly observable quality measure that capital market investors can easily monitor. For these companies, capital market finance is prohibitively expensive.

But capital market borrowing has advantages too. Boot and Thakor’s analysis assumes that private investors can access information about the company’s business that is costly for managers themselves to acquire – for example, about competitors’ behaviour or consumer trends and preferences. These informed traders use the information to change the price of the company’s securities, thereby transmitting that information to the borrower and enhancing the company’s value.

The efficiency with which information is reflected in security prices is directly affected by the skill with which investment banks innovate to produce securities that reflect information. The more they innovate, the more company value is enhanced and the more the moral hazard problem is alleviated, thereby making capital market finance more attractive relative to commercial bank lending.

So investment banks have a clear incentive to innovate and take business from the commercial banks. If, however, the former are owned by the latter, as in a universal banking system, the incentive is naturally blunted because of the depressing effect on commercial bank profitability. The level of financial innovation will therefore be lower in a universal banking system.

But how much lower? The answer depends on the degree of concentration in the banking industry. In a more competitive banking system, the impact of financial innovation by an investment bank on its commercial banking arm will be much more diffuse than when there are few commercial banks.

It is the combination of universal banks and concentration which is most likely to depress financial innovation. And this effect will increase over time as commercial banks become better at monitoring firms. In non-universal or fragmented systems, the opposite is true: the effects of past financial innovations will increase the skills with which investment banks and traders can exploit new innovations. So more competitive and developed financial markets will become more innovative relative to their less developed counterparts.

Which path countries take depends critically on their initial financial structure and the regulatory environment. In countries like the US, where universal banking is prohibited, the market share of commercial banks steadily falls. In countries with a small number of large universal banks, investment banks are likely to find entry very difficult. The commercial banks will have substantial advantages in terms of economies of scale, but they will also have greater incentives to encourage regulators to deter entry by new banks and prevent the development of new financial products.

Boot and Thakor’s theoretical results help to explain the empirical evidence that universal banking has a deleterious effect on financial innovation and the development of financial markets – but only where there is a high degree of consolidation with universal banking, as in Germany. At the same time, there are advantages to universal banks, principally the greater reliance on commercial bank borrowing, which should mean that the moral hazard problem of bad managers exploiting their shareholders is diminished.

So which system should the transition economies choose? These researchers believe that there may be a case for German-style universal banks in Eastern Europe. But, they counsel, simply adopting the universal banking model without being aware of the costs would be a foolish error.