APRA - The New Regime for Prudential Supervision

Jeffrey Carmichael
04 Nov 1998

Introduction

http://www.apra.gov.au/Speeches/Pages/98_10.aspx

Alan Cameron has outlined the new regulatory structure and where the dividing lines are drawn between ASIC and APRA. I would like to begin my talk with some background to why the lines were drawn where they were.

I then intend to add a few comments to Alan's on the way in which ASIC and APRA will work together, then to say a little about the approach we will be taking to prudential regulation, and then to finish with some comments on a topic close to many of your hearts (and bank accounts), namely the cost of prudential regulation.

The New Regulatory Structure

As Alan has mentioned, the new regulatory structure followed from the recommendations of the Wallis Report. That Report recommended a realignment of the then existing regulators, with the result that ASIC gained some consumer protection functions, as well as becoming the sole market integrity regulator, and APRA was formed to integrate the supervisory functions of the old ISC, RBA and the Financial Institutions Scheme.

The basis of these changes was an attempt to realign the regulatory framework from one based along institutional lines to one based along functional lines.

The rationale for this realignment lay in the Wallis Committee's assessment of the underlying motivations for regulation. According to the Committee, the primary rationale for regulation is market failure. In broad terms, markets fail to produce efficient, competitive outcomes for one or more of the following reasons:

• anti-competitive behaviour;
• market misconduct;
• information asymmetry; and
• systemic instability.

All markets face potential problems associated with the conduct of market participants. Anti-competitive behaviour in the form of collusion or exercise of monopoly power has long been recognised as a source of inefficiency in free market outcomes. Similarly, left to their own devices, markets fail as a result of market misconduct in the form of market manipulation and consumer exploitation.

These two forms of market failure are common to all markets, financial and non-financial.

The third form of market failure, information asymmetry arises where products or services are sufficiently complex that disclosure is, by itself, insufficient to enable consumers to make informed choices. To warrant regulation, products characterised by asymmetric information must also involve potentially serious consequences in the event that the promises contained in the product are not upheld.

This form of market failure is not common to all markets. It is present in specialised markets such as drugs and aviation, where products are complex and where promissory breaches can have serious consequences.

It is also present in financial markets. Financial contracts contain promises to make payments at specified times, in specified amounts and in specified circumstances. Not all financial promises are equally onerous and therefore not all financial promises warrant regulation. Financial promises can be distinguished according to the following characteristics:

• the inherent difficulty of honouring the promise;
• the difficulty faced by the consumer in assessing the creditworthiness of the promisor; and
• the adversity caused by promissory breach.

Some financial promises, such as common equity claims, are relatively easy to honour, in that they contain very general and flexible obligations. Other financial promises, such as demand deposits (a promise to pay a fixed nominal amount at the total discretion of the promisee) are very onerous. Similarly, the creditworthiness of some financial promises are extremely difficult to assess - especially where the structure and business of the promisor is highly complex. The consequences of promissory breach can also vary widely. The consequences of a failure of the payments system, for example, would be much more dramatic than the failure of a company to meet its equity obligations.

The Wallis Committee took as a guiding principle that institutions making financial promises warrant regulation only where their promises are judged to have a high intensity in all three characteristics outline above. The Committee saw deposit-taking, insurance and superannuation as all falling into this category and therefore as warranting what we refer to as "prudential regulation".

I should add that the cases for including deposit-taking and insurance were more straightforward than that for superannuation. Ultimately, the case for including superannuation rested on the significance of superannuation to the entire community, the long-term nature of the commitments and the growing involvement of capital guarantees, as well as two Government promises involved: namely, compulsion and tax concessions. There was nonetheless agreement that the intensity of regulation needed for the various promises was different and I will say more on this shortly.

The final form of market failure, systemic failure, is almost unique to the financial markets. It is a fundamental characteristic of parts of the financial system that they operate efficiently only to the extent that market participants have confidence in their ability to perform the roles for which they were designed. Third party, or systemic, risks occurs where failure of one institution to honour its promises can lead to a general panic as individuals fear that similar promises made by other institutions may be dishonoured. Bank runs are the most common example of this type of contagion. However, equally disruptive consequences can also flow from other types of market disturbances such as stock price collapses and even the failure of a single large institution where that institution is involved in a complex network of transactions including forward commitments.

The Wallis Committee suggested that the ideal regulatory structure would be one that had a single regulatory agency dedicated to coping with each of the four sources of market failure. And that is what the Government has given us:

• The ACCC to administer competition regulation - designed to establish laws to prevent anti-competitive behaviour from generating overpricing of products and underprovision of services essential to economic growth and welfare;
• ASIC to regulate market integrity and consumer protection - with the objective of promoting confidence in the efficiency and fairness of markets by ensuring that markets are sound, orderly and transparent;
• APRA for regulating asymmetric information problems - by setting and enforcing standards of prudential behaviour of all institutions making promises in the areas of deposit-taking, insurance and superannuation; and
• The RBA to oversee systemic stability - in particular through its influence over monetary conditions and through its oversight of the payments system.

APRA/ASIC Interface

While this structure involves a much clearer focus for the regulators, it has the consequence that many financial institutions who previously dealt with a single regulator will now deal with two - ASIC and APRA.

As Alan has outlined, in the case of the superannuation industry, ASIC will be concerned with issues such as disclosure and market conduct (including the licensing and behaviour of intermediaries who deal in or advise on superannuation as well as compliance with the non-prudential requirements of the SIS legislation). As he aptly put it: ASIC is concerned with the relationship between the institutions and consumers as individuals. APRA, on the other hand, is concerned with the prudential behaviour of the institution - with its ability to honour its financial promises. This focuses on the relationship between the institutions and their customers collectively.

As Alan has pointed out, the dividing lines between these responsibilities will not always be crystal clear and we will need to work closely together to avoid gaps and overlaps. Alan has already mentioned several of the mechanisms that have been put in place to ensure co-operation between the two institutions. To these I should add that we have recently signed a Memorandum of Understanding outlining, among other things, our:

• Respective responsibilities;
• The need for mutual assistance;
• The establishment of the co-ordinating committee; and
• Information sharing arrangements.

We have also prepared for release a joint APRA/ASIC brochure designed as a practical guide to our relative roles and responsibilities with respect to trustees of corporate, public offer and industry superannuation funds. This guide, which is available at this conference, indicates the role and responsibilities of each regulator.

APRA's Approach to Regulation

Let me turn next to the way in which APRA plans to approach its regulatory objectives.  I should start by stating very clearly that, for the present, our approach to regulation is a simple extension of the approaches of our predecessor institutions. At the broadest level, these approaches are based around the management of risks.

I mentioned earlier the three characteristics of financial promises:

• the inherent difficulty of honouring the promise;
• the difficulty of assessing the promisor; and
• the adversity caused by promissory breach.

Each of these involves risk. The more difficult the promise is to keep, the greater the risk to the consumer. The more complex the institution making the promise, the more difficult it is for the promisee to assess its creditworthiness and therefore the greater the risk. Finally, the greater the consequences of promissory failure, the greater the risk, not only to the individual, but also to the community.

Thus, prudential regulation is largely about managing risk in such a way as to increase the probability of these promises being met - indeed, to increase it to such a high level that they will be met in all but the most extreme of circumstances. Note that this stops short of ensuring that promises will always be met - if this were the objective, the prudential regulator would be effectively guaranteeing the financial promises of regulated institutions.

To minimise the risk to the consumers of financial promises, APRA has to establish and enforce a set of behavioural rules to control the risks that financial institutions run. At present, these rules vary by institutional category. The variations include measurement differences, terminological differences and enforcement differences. The dominance of risk as the focus of regulation also varies across the industry groups.  While these differences are not a source of immediate concern from a regulatory soundness viewpoint, they are a source of concern from the standpoint of uniformity and efficiency.

In my view the biggest single challenge facing an integrated regulator such as APRA is to devise a common framework for analysing all regulated institutions - with the objective that the intensity of regulation produced by that framework for any given institution is a product of the risks that it undertakes, rather than the industry group to which it belongs.  This issue - namely, how to gain the synergies from being an integrated regulator - is focusing the minds of every integrated regulator around the world. Unfortunately there is no magic formula. If there was, someone would have produced it by now.

The following are some thoughts on the ingredients to such an approach. It is meant only to suggest the direction that I hope we can head over the next few years, rather than a clear picture of how we might get there.

In my framework I would break risk into four main categories:

• Credit risk;
• Market risk;
• Event risk; and
• Operational risk.

Every financial institution carries these four risks to varying extents, depending largely on the nature of the promises it makes. For example, deposit taking institutions incur substantial credit risk, some market risk, some event risk (eg., of a liquidity run) and some operational risk.

In the case of insurance, the promises usually contain very little credit risk, some market risk where long-term savings components are guaranteed, but backed by equity investments, very high event risk and some operational risk.

Superannuation involves quite a different risk structure. The primary sources of risk in banking and insurance - credit risk, market risk and event risk - are all borne primarily by the investor in the case of superannuation, rather than by the institution. I say primarily, rather than entirely, because there are some superannuation products (e.g., capital guaranteed products) that leave some of these risk elements with the provider. Consequently, the primary focus of superannuation regulation is on operational risk - managing fraud risk, ensuring good corporate governance and ensuring compliance with the Government's retirement incomes policy.

My point is that risk is fundamental to prudential regulation, but the way in which the regulator goes about managing risk should depend on the nature of the promises involved.

In all cases, APRA's approach is likely to continue to place the primary responsibility for managing these risks squarely on the shoulders of boards, management and trustees.  

The second line of defense will continue to be acknowledgement that boards, management and trustees sometimes get these wrong, and so will require each institution to hold minimum levels of capital and reserves against these risks, particularly where those risks, if realised, are likely to result in losses to the institution's promises. The extent of capital or reserves required, however, should be directly related to the level and nature of the risks involved.

The third line of defense will continue to involve detailed guidelines on the way in which institutions should approach management of certain risks and an inspection process to evaluate the overall interaction of risk management rules imposed by the regulators and those imposed by boards and management.

The Cost of Regulation

Let me turn lastly to the much vexed issue of the cost of regulation.

Let me start by saying that the process was rushed this year and not to anyone's satisfaction. Next year, there will be a much more extensive process of consultation with industry before a recommendation is made to the Treasurer.

That said, I should add that you should not hold your breaths for a significant decrease in your regulatory levies in the near future.

The general parameters governing the cost of regulation by APRA, as laid down by the Government, are as follows:

1. There has been no increase in the total cost of on-going prudential regulation (indeed, our total operating budget has been frozen in nominal terms in forward budget projections);
2. We are required to fully recover the costs associated with regulating each industry sector - I should add that the recovery includes the cost of consumer functions passed to ASIC plus the ATO's costs associated with unclaimed monies and lost member functions; and
3. We are required to recover from each industry group a share of the establishment costs of APRA.

Notwithstanding these parameters, levies for each industry are different and the change in levies has varied greatly across industry group. Let me offer a few comments about some of the apparent anomalies:

The most fundamental point is that all industries have experienced (or, will experience) a big change in their levies because of the establishment costs and because past levies bore little relationship to the actual cost of regulation:

• Banks, for example, will pay much reduced levies from 1999 because their old (implicit levy) structure - related to a penalty interest rate on NCDs - was many multiples of the actual cost of regulation;
• While 85% of superannuation funds experienced no change in their levy in 1998/99, the remaining 15% experienced an increase. This was due to a number of factors:

1. Most importantly, past levies have underfunded the operating cost of regulating non-excluded superannuation funds.

2. The underfunding by non-excluded funds has been masked by a substantial cross-subsidy from excluded funds (which currently only account for around $1.5 million of regulatory costs but contribute close to $30 million to regulatory levies). In the past, not all of the levies flowed to the regulator and around $9 million per year of this will disappear in 1999/2000 and beyond.

3. The 1998/99 levies included an allowance for the establishment costs of APRA - and these will continue for the next four years.

4. Finally, the change in basis of levies from a fixed rate per tranche of assets to a straight percentage levy on assets, combined with an increase in the maximum levy payable, changed the calculation basis for levies and increased the cost for large funds.

While the shock of paying more takes a little getting used to, the amount paid is still relatively small and, I suggest, with a maximum levy of $39,000, is unlikely to plunge any of you into financial difficulty. I should, however, in wrapping up, leave you with some food for thought:

• First, that cross subsidy will disappear next year. Be prepared for an increase in levies.
• Second, it is our committed intention to gain some cost efficiencies over time from our structure as an integrated regulator. While this should lead to a reduction in the real cost of regulation over time, it will not happen overnight.
• Finally, the approach I outlined earlier should involve a substantial change in the way superannuation is regulated. This is likely to lead to a reduction in the cost of regulating superannuation relative to the cost of regulating other financial promises. Again this is not going to occur over night, but it is a serious medium-term objective.