Reading Amar Bhide's new book made me wish that more people shared my outlook on banking, finance, and financial regulation. I think it helps to keep in mind the following (continued below).
1. In the nonfinancial sector, people want to issue long-term, risky liabilities (shares in fruit trees) and to hold short-term riskless assets (bank deposits). Call these L's (long-term) and S's (short-term), respectively.
2. Financial intermediation attempts to transform some L's into S's. Think of the financial sector as holding L's as assets and S's as liabilities, so that the nonfinancial sector can do the opposite. Banks are financial intermediaries, but nonbank firms are financial intermediaries also. The liabilities of nonfinancial firms (debt, loans, commercial paper) are more S-like than their assets (organizational capital, physical capital).
3. Financial intermediation uses three tools to add value and create profits. One tool is hands-on due diligence. Another tool is diversification. The third tool is confidence-building.
4. Hands-on due diligence means that the bank spends effort learning about the fruit tree projects before selecting the projects in which to invest. It also monitors the fruit tree managers. The bank has skills and information that its depositors lack in undertaking these tasks. In a sense, depositors are paying banks to undertake these due diligence efforts. This means that we are not in a Modigliani-Miller or Arrow-Debreu world of transparent connections between individuals and the projects in which they are invested. People who put their savings into financial intermediaries (including the securities issued by nonfinancial firms) are doing so with imperfect knowledge of the underlying risks.
5. Diversification is the pooling of risks. Rather than forcing depositors to bet on one fruit tree, the bank assembles a diversified portfolio of fruit trees. Rather than relying on a single depositor, the bank uses many depositors. Because not all of them will need to access their funds the same day, the bank can safely issue S's and hold L's. Note that while due diligence is hands-on, diversification per se is hands-off. It relies on probability distributions rather than specialized knowledge. Amar Bhide seems to say "hands-on good, hands-off bad," but in fact both are useful.
6. Confidence is important, because you only want depositors to withdraw funds when they really need the funds. If they withdraw funds because they are afraid the bank is in trouble, the bank cannot safely issue S's and hold L's.
7. The importance of confidence creates the Minsky problem. That is, stability creates instability. As people become more confident, financial intermediaries find it easier to take more risk. At some point, the risks become excessive, and you get a crash. Then the challenge becomes getting intermediaries to take risks again. Confidence is inherently cyclical.
8. The importance of confidence also creates the temptation for government to step in with guarantees and with regulatory attempts to prevent crashes. Government acts as if the Minsky cycle can be conquered.
9. Government guarantees and regulatory regimes have their own life cycle, which perhaps we could call the Kling problem. The regulatory reforms that are enacted in response to a crisis could indeed prevent the exact same crisis from recurring. However, over time the financial system responds by adapting to game the system, putting an increasing share of risk on the taxpayers. This gaming takes place through financial innovation and political lobbying. Eventually, the regulatory regime breaks down, another crisis occurs, and the taxpayers get stuck with a large bill. New reforms are enacted, and the cycle begins again.