Conference – 1996 Discussion

1. Vince FitzGerald

Overall, this is an excellent background and discussion opening paper for this conference, presenting:

  • a concise chronology of relevant changes in the Australian financial system;
  • a well-selected portfolio of quantitative exhibits, in the familiar RBA style, showing some of the major trends; and
  • a good discussion seeking to identify some of the drivers of change.

I have only a few points to make.

First, I think the paper could have brought out more strongly what a challenge the financial sector is becoming to taxonomers. We are increasingly – and correctly – talking about it in the same terminology we apply to other services markets – that is, in terms of products or services meeting particular customer needs, yet we have not traditionally collected data in these terms. No doubt this is partly due to the joint provision and joint pricing practices which so long prevailed – obviating the need for providers (such as banks) to maintain data on the output or volume, costs and pricing of separate financial services. Rather we collected data on financial stocks and flows, mainly in institutional categories. These categories looked fairly stable in the past but are plainly less stable now than the overall demand for the generic products.

For example, where can one find data on the markets for deposit services or transactions services? What prices are being received for what quantum of these services supplied by what providers to what customers? Assets, liabilities and even turnover data for types of institutions do not provide such measures. Banks may have dominated these areas historically, but what about the non-banks which were and are providers of essentially the same kinds of services (for example, building societies and other ‘intermediaries’, in the sense of this word adopted by Edey and Gray, many of which later became banks)? What does Figure 2 convey other than that some long-standing players in this market, who weren't formerly in the bank ‘club’, now are? Should we now also count, for example, payroll companies, IT companies like Microsoft or Intuit which are beginning to do transactions business in the United States and touching our market and others via the Internet? Telstra and Optus in the near future?

The paper does say at the outset that it will raise some questions about the boundaries between traditionally defined institutions which form the basis for existing supervisory/regulatory structures, but in fact it seems to accord these institutional boundaries – and indeed existing policy and regulatory structures themselves – considerable respect throughout and, if anything, goes out of its way to defend their continuing relevance. Two examples are the statement in Section 2.3.3 that ‘household behaviour seems to make a clear distinction between deposits with intermediaries and balances with funds managers’ and the one in Section 4 that ‘a basic distinction in principle can be made between intermediaries which offer deposit and loan services on a capital-guaranteed basis, and funds managers, which manage but do not bear investment risk’.

  • In respect of the former quotation, it is clear that in the marketplace for accumulation savings products, term deposits are not a clearly distinct species, but sit closely beside such alternatives as debentures, annuities, unit trusts and life and friendly society bonds. The broad layers in Figure 4 tend to bury rather than reveal that there is this range of substitutes crossing the institutional boundaries.
  • In respect of the second quotation, there is surely no ‘basic distinction in principle’ between deposits and at least some of the products of the funds-management sector (as defined by Edey and Gray), but rather a spectrum of sharing of risk. A capital-guaranteed life insurance product can be generically virtually identical with a term deposit; and equally relevant, both may be offered by diversified financial services groups, whether based around established ‘banks’ or established ‘life offices’.

It does still seem true in Australia that the cultures are different as between ‘bankers’ and ‘life insurance officers’ or funds managers, but these cultures are being blended in the 1990s in most major diversified financial services groups, not just those based around established banks. Again, what we see is a progressive blurring of institutional distinctions.

Turning to another general comment, I think that in the explanations of factors driving change which are given in the paper, a little too much weight is given to policy and regulatory factors relative to the more fundamental or underlying (mainly economic) forces at work, which might loom larger if we took a more generic (rather than institutional) view of the financial sector.

A case in point is the growth of superannuation – a species of the generic class of accumulation savings products. In respect of superannuation, the discussion in the paper focuses heavily on the role of policies, especially the policy imposing compulsory minimum contributions, and major tax policy changes – while at the same time trying to explain why (despite compulsion) net contributions to the system have so far apparently accounted for so little of the growth in assets. I note in passing that the validity of the available data on net contributions is the subject of considerable debate among those familiar with these data and their sources. But my main point is that Australia's rise in this type of financial wealth accumulation is not very different qualitatively from that in the United Kingdom or the United States, for example. This similarity of trends is even clearer if the view is broadened to include not only pension or superannuation funds but mutual funds, unit trusts and so on. But in any event it certainly applies in respect of pension or superannuation funds alone, despite our very distinctive policies. Therefore it seems more likely that there are underlying common factors across these countries other than policy or institutional factors – for example, demography, or the long postwar period of rising overall personal wealth (suggesting a rising proportional allocation to longer-term financial accumulation products, typically invested in marketable securities) – but not much attention is given to these sorts of factors in the paper.

I note in passing that I think that the point in Section 4.1 about increased tax rates discouraging voluntary contributions to superannuation is incorrect. The incentive to make such contributions, particularly for upper income earners approaching age 55 (when they can access the funds) is still extremely strong – a shelter of over 33 percentage points at the point of contribution.

I believe further that Australian households are still able to target the overall shape of balance sheet they want – in broad generic terms – with so far only a modest effect from compulsory superannuation policies. Ability to leverage using dwellings as collateral is one obvious means that households are using to offset such policies. In this regard, I do not think that looking for competition for flows of new saving is the sole place to look for offsets to compulsory superannuation saving. Studies focusing on short-term substitution between flows find that the apparent offset is relatively small, but miss the bigger balance-sheet adjustments – that is, that households have increased their use of debt to the extent that they are now cashflow negative with banks, and have continued to reduce net saving rates steadily over the whole decade since award superannuation (the precursor to the Superannuation Guarantee) was initiated.

Finally, I agree with the distinctions drawn in the paper between the distinct businesses, activities or services (deposit-taking, lending, transactions etc). However, especially since we are now seeing any or all of them offered by providers with different institutional histories, I wonder whether we should not now de-emphasise the old institutional distinctions and concentrate on the functions themselves – as products or services which virtually any financial services firm (or group) can now offer.

In the Wallis Inquiry context, one implication is that specialisations in regulatory activity should perhaps in the future be organised around generic or functional categories of business (for example, payments services or deposit-taking or life insurance), regardless of what kind of financial services group offers them. And we clearly need to develop new views of how risks aggregate to the level of a financial services group as a whole from its various businesses, and indeed of what may now give rise to ‘systemic risk’.

2. General Discussion

The discussion centred on the competitive pressures facing banks and their impact on pricing structures and profitability.

A key issue was that of ‘unbundling’ – the process whereby competition in banking was developing at the individual product level rather than on a full-service basis. This was putting pressure on banks to price each individual product more competitively, to reduce margins on the most profitable lines, and to cut cross-subsidies. Participants debated how far this process was likely to go.

In discussing this issue it was noted that analysis was hampered by a lack of relevant data on costs and prices at the product level. The information produced by banks has tended to be highly aggregated, with costs and prices averaged across a wide range of banks' activities. Some participants remarked that this style of reporting reflected the way banks themselves have traditionally thought about their operations: they have been concerned with overall market shares and with the average profits of their total operations, rather than being focussed on individual products and markets. It was remarked that this approach would have to change, since the main competition for banks was coming from the specialist service providers.

The vulnerability of banks to specialist competition would depend importantly on the extent to which their existing pricing structures involved cross-subsidisation: new entrants would target the most profitable products, which were the source of revenues for any cross-subsidies. The rise of mortgage managers was an important case in point.

Some participants took issue with analysis in the paper which concluded that there were significant cross-subsidies built into banks' traditional pricing structure. In particular it had been argued in the paper that loans and deposits tended to be priced on an average-cost basis, where the average margin cross-subsidised the provision of transaction services. Participants who disputed this pointed to a distinction between cross-subsidisation and price discrimination. The latter, which involves tailoring products and prices to the individual customer, was argued to be quite sustainable even in a competitive market, and could be viewed as a normal way for banks to recover fixed costs which could not be directly attributed to an individual product. It was argued that it can be hard to tell the difference between this sort of behaviour and cross-subsidisation without detailed information about the sources of banks' costs. Since this information is not available, it was argued that we should be cautious in drawing conclusions in this area.

Other participants argued that a strong element of cross-subsidisation was occurring. They felt this view was consistent both with Australian evidence and with experience overseas. One comment was that the pricing of bank services had been strongly driven by public pressure on the banks – particularly the resistance to higher transaction charges. But competitive forces would inevitably shift the industry, in time, towards a more rational pricing structure. Indeed, even if the ‘price discrimination’ view outlined above was accepted, the increasing sophistication of customers was likely to have a similar effect on the prices they could charge. The net effect would be a squeezing of margins on the most profitable of the banks' products. Another factor reinforcing this trend was the shift to low inflation. This meant that average nominal interest rates had fallen, and banks could no longer recover costs from low-balance high-transaction customers through the interest margin.

Two consequences of these developments were discussed. The first was that banks faced increasing pressure to charge more for underpriced services, particularly for transactions. This in turn might mean a more open market for transaction services with new entrants being attracted. If this were to occur, it would reduce the rationale for special regulation of banks, to the extent that such regulation was motivated by banks' special role in the payments system. The second was that bank profits were likely to come under downward pressure and that the returns on capital of the order of 15–20 per cent seen in the past could not be sustained.

There was also some discussion of trends in the superannuation sector. It was noted that high rates of return in recent years would have contributed to reductions in voluntary contributions where defined-benefit schemes are concerned, since less contributions would be needed to fund the final benefit. Even though defined-benefit schemes are no longer the norm, this sort of effect might still be important for ‘target’ savers making voluntary contributions.