Why Do Banks Need a Central Bank?

Summary

A Central Bank has two main functions. Its first (macro-economic) function, the operation of discretionary monetary policy, has already been touched on in Chapter 6, and will be further discussed in Chapter 10 and Chapters 13, 14 and 15. In this chapter, by contrast, we consider a Central Bank’s second (micro-economic) function, of providing support (e.g., via Lender of Last Resort assistance), and regulatory and supervisory services to maintain the health of the banking system.¹

One of the reasons why many markets work imperfectly is that there are differences in the information available to the various participants in the market — i.e., information asymmetries. As noted in Section 1, this feature is particularly marked in the case of the provision of professional services, including investment and financial advice, but extending far further, encompassing doctors, lawyers, etc. Since what is being sold in such cases is, in some large part, specialised knowledge, information asymmetry is inherent. Furthermore, an individual’s requirement for such assistance is usually infrequent, but the potential impact on their utility if the professional gets it wrong can be devastating. So, the usual guarantor of quality, which is the need to protect reputation to ensure repeat buying, may be insufficient. In such cases, the professionals involved may come together to form a ‘club’, to provide entry controls and guarantee quality, though often also to restrict competition and raise minimum charges, etc. Some facets of a Central Bank’s micro-function may be seen in its adoption of the role of manager of the ‘club’ of banks.

But the role of a Central Bank, in its support for the banking system, goes much further than that of a typical central professional body — e.g., in regulations on, and lender-of-last-resort support for, domestic commercial banks. It is often argued that the need for a Central Bank to carry out such functions arises from the joint role of banks in providing both portfolio management and payment services. Thus, if the portfolio management turns out badly, and losses are made, the public good of the provision of payments services throughout the banking system may be put at risk. This claim is analysed in Section 2 and it is argued that it is not generally valid. It depends on the particular characteristics of the typical asset portfolio held by banks. A number of suggestions have been put forward to restrict the range of assets banks could hold. If banks, or other intermediaries offering payments services, were restricted to holding absolutely safe assets, or were restricted to holding only marketable assets, and were closed as soon as a fall in the value of such assets reduced the positive capital value (solvency) of the banks to some preordained limit, then the support role of the Central Bank could be much reduced. In Section 2, I offer yet another alternative suggestion, that financial intermediaries might offer payment services on the basis of liabilities whose value might vary in line with the value of its marketable assets. Once again, such an institution should be safe, and protected against runs, so Central Bank support would be otiose. Whether such a development is likely to take place is considered.

But if banks were restricted to holding only marketable assets, who would provide the non-marketable loans which now provide the staple of bank assets? In Section 3, I describe how banks have a comparative advantage in the information, monitoring and enforcement arrangements inherent in the exercise of making loans, when the alternative of borrowing through primary markets is too expensive, primarily because of the information costs necessary before primary markets can work efficiently, together with other causes of economies of scale therein. The nature of the involvement between bank and borrower has led banks to make such (non-marketable) loans on a fixed-nominal-value basis; similarly, the nature of the relationship between depositor and bank has caused bank deposit liabilities to be on a fixed-nominal-value basis.

But this condition, in which banks make non-marketable loans, of an unknown ‘true value’, on a fixed-nominal-value basis backed by liabilities also convertible into cash at a fixed-nominal-value, brings with it possibilities for abuse by bankers and for contagious panics. Even if banks did not offer any payments services — i.e., if they funded themselves only with time deposits or CDs — these circumstances would still involve the need for bank regulation, and for Central Bank support to prevent crises affecting sizeable parts of the banking sector — i.e., systemic crises.