While cooperatives in many developing countries provide deposit services (and other financial services) to large numbers of low- and middle-income clients who often cannot access banks, they have typically not been subject to the same kind of regulatory scrutiny as banks. Cooperatives lack this scrutiny because they tend to hold a far smaller fraction of system-wide assets, and they are often considered more "cooperative" than "financial" institutions. However, given their importance in providing deposit services to large numbers of low-income clients, improving their regulation and supervision is currently a topic of intense debate. Below, Carlos Cuevas of the World Bank's Financial Sector department shares his views on how best to protect the savings of cooperative members.
1. What are the arguments for and against R&S of CFIs?
The arguments for (external) R&S of CFIs are not too different from those for the regulation and supervision of banks. Governments regulate the solvency of financial intermediaries to protect depositors and to foster financial system stability. Although CFIs as a sector will rarely matter greatly in terms of financial system stability--as they usually account for a small fraction of total deposit amounts in a given country--CFI depositors are still totally worthy of protection. One could argue even more so, as they are typically low- to middle-income households whose savings with CFIs may have more the nature of an emergency reserve than one of speculative investment.
For relatively small CFIs, costly external R&S would not be justified. Instead, well-functioning internal control mechanisms (such as a "monitoring/control committee") may be sufficient to ensure depositor protection and to prevent management and boards of directors from exposing depositors' funds to unwanted risks. As CFIs become larger than a "critical scale"--i.e., one in which individual members have reasonable knowledge of (and trust in) those serving in the governance bodies--internal controls may fail to prevent board or managerial abuse of decision-making powers; hence the need for external R&S.
The arguments against R&S of CFIs are typically associated with the costs of supervising large numbers of relatively small institutions. In other words, opponents do not deny the need to protect depositors, but they claim that it is too costly. In my view, the question, therefore, is how to set up a supervisory system that minimizes those costs while still providing credible protection.
2. How do principal-agent conflicts differ in CFIs and investor-owned financial institutions? In CFIs, is the member-manager conflict more important than the net saver-net borrower conflict?
In an investor-owned bank the main agency conflict is between shareholders (owners) and depositors; owners are willing to take risks with the depositors' money that the depositors may deem excessive. In the CFI, owners and depositors are the same people, hence there is no shareholder-depositor conflict. The member-manager conflict is the main agency conflict, as managers may make expense decisions that debilitate the institution. Some experts argue (such as my colleague Brian Branch of WOCCU), and I agree, that in many instances it is not the manager but members of the Board of Directors (or other governing bodies) of CFIs who engage in or induce expenses or risk exposure that go against the members' interests and safety.
On the net saver-net borrower conflict, our work indicates that, while this conflict should not be ignored, in practice it is less significant than theory and limited evidence suggest. "Borrower bias" has been known to occur in environments of little competition and subsidized sources of funding (as in Latin America in the 1960s) which also aggravates the member-manager conflict. Hence, observed failures may well be a combination of both types of conflict. Interested readers may want to look at the paper Klaus Fischer and I have written (Cooperative Financial Institutions: Issues in Governance, Regulation and Supervision; World Bank Working Paper 82, July 2006) where we cover these conflicts extensively.
3. What type of R&S most effectively minimizes both the risks inherent in CFIs and the costs of supervision? Why?
This is a tough question; it invites the usual "it depends" answer. The answer is in fact, in my opinion, a pragmatic one which should take into consideration the scale and diversity of the CFI sector in question. If the authority intends to protect, say, at least two-thirds of all depositors in the CFI sector, the practical question is "how many CFIs need to be supervised to reach that target?" In some countries it may be a handful of institutions (e.g., Kenya), or even a dozen, in which case a direct supervision approach by the financial authority would be highly effective and not overly costly. The problem emerges when to reach that coverage target we are talking about a couple of hundred institutions. Then a so called "tiered" system that only supervises the largest CFIs may end up protecting a minority of all depositors. In this situation, indirect supervision mechanisms may be the best compromise between effectiveness in risk reduction and costs of supervision.
4. How can the problems associated with indirect supervision be minimized?
Indirect supervision mechanisms typically rely upon the existence of CFI networks or alliances (federations or other second-tier organizations) with which the regulatory authority would need to work to exercise its powers. Problems in indirect supervision are likely to emerge when those networks have weak governance, when they are dominated by a large and weak CFIs, and when there is a multiplicity of roles (such as advocacy and promotion along with that of supervision) without the requisite "firewalls." A first concern, therefore, is to ensure the quality of the second-tier entity (alliance, network, or federation) in terms of governance, composition, and functioning of governing bodies. This can be done through legal clauses and regulation of governance functions (e.g., rotation and "fit and proper" provisions for board members) which unfortunately are still rare in CFI legal and regulatory frameworks. Clearly separating the supervisory unit of the network (i.e., that which cooperates as an auxiliary arm of the regulator or through explicit delegation of certain powers) and ensuring that members of the supervisory unit are certified by the regulator and autonomous (vis-à-vis the governance of the network) are other means of mitigating the risks of conflicts of interest in indirect supervision. Systems that operate with indirect supervision (such as the German and the Quebec systems) have fully professional, "fire-walled" supervision units at the network level. In developing countries, the Mexico system is following this model. Making the auxiliary supervision bodies cover more than one network is an additional way of minimizing conflicts-of-interest risks.
5. What further research is required before a common guideline for CFI R&S can be established?
It seems to me that most of the additional research needed has to do with identifying the conditions under which alternative R&S models are more effective than others (including cost effectiveness). Cross-country analyses of CFI performance, relating performance with industry structure and R&S systems, would take us a long way towards defining guidelines which are coherent with the scale, diversity, and state of development of the CFI sector. How important is integration (networking)? How should CFI R&S fit within the same authority that regulates and supervises banks? These are some of the key questions that concern most stakeholders.