RDP 9315: The Provision of Financial Services – Trends, Prospects and Implications 3. Developments to Date

3.1 International Experience

3.1.1 The Corporate Sector

The structure of corporate balance sheets varies considerably across countries (Figure 1).^[9],^[10] Leverage is very high in Japan and Germany and relatively low in the USA, the UK and Canada.

While many factors influence leverage, international studies find that an important explanation for the difference in leverage across countries is the extent and nature of information flows between firms and the financial institutions in each of the countries.^[11] In those countries where leverage is high, there is a relatively close relationship between banks and the corporate sector. Banks can hold significant amounts of equity in companies they lend to, exercise voting rights at shareholders' meetings and have representatives on firms' boards.^[12] They thus have access to detailed information on the firm. This information may not be widely available as reporting standards in these countries do not require extensive provision of information on firms.^[13] The relationship between banks and their corporate clients is at ‘arm's-length’ in the low-leverage countries, where corporate disclosure is also more stringent.

By reducing the uncertainty surrounding a firm's ability and willingness to repay a loan, the close relationship between banks and the corporate sector in some countries also enables firms to be less reliant on internal sources of funding (Table 1). Cash flows tend to be a more important source of funds in countries with arm's-length banking (the US, UK and Canada) and less important in the other countries. Arm's-length relationships and the wider dispersion of corporate information also encourages greater use of funding through securities markets. Funding through securities markets – both debt and equity – are a more important source of new funding vis-a-vis bank lending in the US, the UK and Canada (Table 1).

These data highlight the importance of how the resolution of information asymmetries between borrowers and lenders can shape the financial system and corporate balance sheets. As Bisignano (1990) points out:

‘The market for finance is inseparable from the market for information… The modest public availability, close sharing and harbouring of information by firms is at the source of the greater reliance of continental European and Japanese companies on intermediated finance and for the greater use of debt in business financial control… The … bias in Anglo-Saxon countries is for a wide dispersion of business information and corporate ownership claims and a competitive market in corporate control.’^[14]

The influence of these factors on financial structure is clearly not immutable as evidenced by the shifts in financing over the past decade or so. There has been some convergence in gearing across countries (Figure 1). Leverage rose in the traditional low-leverage countries and fell in the high-leverage countries. In many countries, corporate indebtedness grew more rapidly than the overall economy (Figure 2). An important feature of this data is that bank lending to the corporate sector was a relatively stable and, in some cases, large component of corporate balance sheets.

A significant part of the rise in corporate funding in all countries was used to acquire financial assets. Growth in investment in financial assets outstripped the growth in investment in non-financial assets over the whole period, reflecting the increased financial sophistication of firms, the growth in the range of financial assets and the high returns available on financial assets relative to funding costs at the time.^[15]

3.1.2 The Household Sector

Over the period since the mid 1970s, household assets have grown more rapidly than GDP in all countries – almost doubling in Japan and growing by some 40 percentage points of GDP in the other countries (Figure 3). Growth has been least significant in the US, where the ratio of assets to GDP is far higher than in any of the other countries. In all of the countries except Japan, almost all of this rise has been due to an increase in funds held outside deposits at intermediaries. In many cases, household holdings of equity, either directly or through mutual or pension funds, increased.

While there has probably been significant discretionary portfolio reallocation going on in household balance sheets, some part of the rise in the share of assets held in equity, mutual and pension funds probably reflects the sharp rise in asset prices and returns on these investments over the 1980s. This would help explain why household financial assets rose sharply over the decade in most countries while household saving ratios fell in most countries except Japan (where they were flat) and the UK (where they rose sharply at the end of the decade).

An important feature of these data is that while the share of deposits in the household sector's portfolio declined, they remained stable relative to GDP in most countries indicating that banks remain an important repository for household financial wealth.

The increase in financial wealth in conjunction with rising house prices during the decade boosted the collateral of households and encouraged them to increase their indebtedness (Figure 3). The introduction of home-equity loans and the general liquidity enhancing nature of many innovations enabled households to support higher debt burdens. Since most of this has been funded by traditional banking institutions it has been of direct benefit to the banking sector. To some extent, this may have compensated the banks for their loss in market share of household assets over the period.

3.1.3 Financial Institutions

How did these changes affect banks?

The two most important observations worth noting are that there has been an overall increase in household and corporate balance sheets in most countries and significant shifts in the structure of both the asset and liability sides of those balance sheets. Analyses of the performance of banks almost universally focus on the second point without considering the first, painting a very pessimistic picture of their performance.^[16]

While banks have become a smaller part of the financial system, they have benefited from the overall expansion of the system. With the exception of the US, bank assets have grown much more rapidly than the rest of the economy in the major countries (Figure 5). Even in the US they have been stable relative to GDP.

There have been changes within banks' assets. Corporate lending has become a smaller part of banks' asset portfolios. This reduction probably stems mainly from a shift by some large firms to non-bank sources of finance. Available evidence suggests that small firms and relatively new firms remain as reliant on bank loans as a source of external funds.^[17] Bank loans to the corporate sector remain, nonetheless, an important source of corporate funding needs (Figure 2) and an important component of banks' balance sheets.^[18] Non-deposit liabilities have become a more important source of funding for banks (Figure 5). Banks have increasingly borrowed from the securities markets to fund their expansion.^[19]

The fact that bank assets have expanded more rapidly than the rest of the economy suggests the continued importance of banks in the financial system. Indeed, growth in the assets of the banking sector substantially underestimates the growth in total banking activity since it excludes banks' off-balance sheet activities. Off-balance sheet items including loan commitments, standby letters of credit and the provision of derivative instruments (options, futures and swaps), have grown rapidly in the 1980s. The growth in banks' off-balance sheet business is reflected in the growth in net non-interest income as a share of gross income of banks (Table 2). Growth in net non-interest income has occurred in all but one of the G7 countries over the 1980s as banks, in the face of increased competition for their traditional business, sought other sources of profit and competed directly with the securities markets.

The growing use of direct financing in the securities markets by firms has been one of the important new sources of competition facing banks. This has largely involved the movement of high-quality components of lending – mainly large low-risk commercial loans – off banks' balance sheets. There are limits to this process, however, as the idiosyncratic nature of some loans makes them difficult to price in securities markets. Moreover, evidence suggests that many of the funds being channelled through securities markets and finance companies are intermediated. In the US, around 90 per cent of commercial paper issued by the largest finance companies is backed by banks (Edwards 1993). And, the majority of securities backed by residential mortgages have direct government guarantees.

Despite the expansion of banking activity, the growing competition banks have faced has resulted in a fall in their profitability (Table 2).

3.2 The Australian Experience

3.2.1 The Non-Financial Sector

Structure

Households and firms in Australia are funded through a range of institutions and instruments. Banks provide over 70 per cent of household finance with most of the remainder coming from non-bank deposit taking institutions.

Banks also provide the bulk of the corporate sector's debt finance either directly through loans (26 per cent) or indirectly by holding bills (6 per cent) (Table 3).^[20] Non-bank financial institutions provide about 20 per cent of corporate debt funding with a further 20 per cent being raised from the rest of the world. Life offices and superannuation funds provide very little debt finance. The largest direct holder of corporate equity is the foreign sector. Domestic households directly hold about 25 per cent of corporate equity and indirectly another 20 per cent through life offices and superannuation funds.

Trends

The evolution of the financial assets of the household sector in Australia was broadly similar to that in other countries. They grew very rapidly during the 1980s, with most of the growth occurring through increases in household claims on life offices and superannuation funds (compare Figure 3 and Figure 6).^[21] In Australia, and perhaps elsewhere, the growth in assets held in superannuation funds was primarily due to the high earning rates achieved by superannuation funds. On average, net contributions to the funds were little different to what they have been in the past thirty years.^[22] Household holdings of equities and units in trusts also grew strongly. This was in response to the higher yields available on these assets for much of the period. Household deposits in financial intermediaries – banks and non-banks – have declined as a proportion of household sector financial wealth. Over the 1980s, deposits at banks grew at the expense of deposits at NBFIs. In most countries, deposits at banks grew at least as rapidly as the rest of the economy despite the increased competition faced by banks.

The indebtedness of the household sector also rose during the 1980s (Figure 7). Like the other countries, the increase in household debt was less than that of household assets. Household debt in Australia seems low in comparison to several of the major countries.

Corporate balance sheets expanded very rapidly during the 1980s (Figure 8).^[23] Internal funds were an important part of this expansion, reflecting the recovery in the profit share over the decade. Both sources of external funding – debt and new equity raisings – also grew quickly but, over the course of the decade, there was a shift towards greater reliance on debt funding. Corporate gearing increased sharply but remained relatively low compared with international experience (Figure 9)^[24].

3.2.2 The Counterpart Experience in the Financial Sector

How did these changes in the size and structure of the private sector's balance sheets relate to the behaviour of financial institutions?

The 1980s was a decade of financial deepening. There was a rapid expansion of the assets of the financial institutions, which have grown by 86 percentage points of GDP since 1980 (Figure 10), following the repressed financial behaviour of the 1970s.^[25] The increase in the volume of activity was spread between most of the major institutional sectors, in particular the banks and the life offices and superannuation funds.

Looking behind the aggregate data it appears that intermediated funding – through banks and NBFIs – provided the bulk of the increase in external finance to the private sector (Table 4). New funds raised by the private sector rose from the early 1970s, peaking in the second half of the 1980s. Intermediated funding progressively became the most important source of funding. Within this total, bank finance – traditional lending and bills – grew very rapidly as banks won back significant market share from NBFIs. This increase in intermediation in general, and in bank finance in particular, reflected the removal of regulations on bank behaviour.^[26] Over the past few years, the repayment of debt and new equity raisings by companies has seen the debt component of new funding fall.

A feature of the Australian system is the limited use of debt raisings through the securities markets. Their direct contribution to the increase in corporate funding was negligible over the 1980s. Long-term debt securities and $A Eurobonds account for only 2.8 per cent of outstanding corporate debt. $A Eurobonds outstanding grew from nothing in 1984 to $25 billion in 1993. The growth in Eurobond raisings corresponds with the growing reliance of the government, finance and corporate sectors on overseas finance. While these markets quantitatively are a small funding source, qualitatively they probably had a more important indirect influence on funding and the financial system by increasing the competition for traditional sources of funding.

The following sections will explore in more detail how these changes affected bank behaviour and profitability.

Bank Balance Sheets

Deregulation and the deepening of financial markets allowed banks to compete more effectively for deposits and to obtain new sources of finance to directly fund the growth in private sector debt.

Deregulation allowed banks to adjust interest rates on deposits more closely with money market rates (Figure 11). As a result, they could actively manage the liabilities side of their balance sheet, enabling them to more effectively fund the expansion in private lending. A consequence of their growing competitiveness and flexibility was that bank-owned subsidiaries, previously set up to avoid the effects of some regulations, became incorporated into the banks and some building societies changed to banks. A trend decline in the bank share of the deposit market was halted.^[27] Household deposits at banks rose largely at the expense of deposits at NBFIs. Despite the increase in the relative return on bank deposits they did not rise as a share of the household sector's portfolio (Figure 6).

Nor did household deposits fund the bulk of the rise in bank liabilities (Figure 12). One reason for this was that higher deposit rates probably just let banks hold their own in the market for household funds rather than significantly alter the allocation of household financial assets. More importantly, until 1988 the Statutory Reserve Deposit (SRD) requirement discouraged banks from funding loans through deposits.^[28] Banks could avoid the implicit tax of the SRD requirement by raising funds through foreign currency deposits (which they were unable to do prior to 1984), issuing their own bills or equity. These other sources of funding provided the bulk of the rise in bank liabilities over the 1980s (Figure 12). The removal of the SRD requirement in 1988 removed this disincentive and there has since been a shift towards deposits and away from other sources of finance (See Reserve Bank of Australia 1989).

As a result of these changes, bank liabilities pay closer to market returns and a significant portion of bank funding is provided through domestic and international financial markets. This shift to non-deposit sources of funding is also evident in other countries (Figure 5).

Bank assets rose in conjunction with the increased indebtedness of the household and corporate sectors (Figure 13). Lending to households by banks rose from 12 per cent of GDP in 1980 to 32 per cent of GDP in 1993. At the same time, on-balance sheet lending to firms increased sharply. Bank lending to corporates stood at 18 per cent of GDP in 1980 and 35 per cent of GDP in 1993.

Off-Balance Sheet Activity

The balance sheet data presented above understate the volume of banking activity. Off-balance sheet activities have become an extremely important source of income for banks. As at June 1993, gross off-balance sheet business was around 6 times banks' consolidated balance sheet assets. Measured on a credit equivalent basis, off-balance sheet activities amounted to 29.7 per cent of bank assets.^[29]

While consistent data is unavailable over an extended period of time it is safe to say that the volume and variety of off-balance sheet business grew rapidly in the 1980s (see Reserve Bank of Australia 1991). Banks have always been engaged in off-balance sheet business such as letters of credit and lending commitments but began to extend into market-rate related transactions – foreign exchange transactions, swaps, options etc – as markets for these instruments developed in the 1980s. They now make up the bulk of banks' off-balance sheet business.

The expansion of banks' off-balance sheet activities has seen a greater proportion of their income coming from fees rather than interest income (Table 5). This is a feature common to most countries.

Table 5 : Australian Banks Net Non-Interest Income as a Share of Gross Income

1986	39.0
1987	42.9
1988	39.4
1989	38.1
1990	40.1
1991	47.7

Bank Profitability

As demonstrated above, a significant part of the increase in the volume of financing during the 1980s was intermediated through banks, either on or off their balance sheets. To win this increased business, however, banks had to compete more vigorously on both sides of their balance sheets. This, in part, could have contributed to an erosion of bank profitability as the returns previously earned by bank shareholders in the regulated environment were partly dispersed to depositors and borrowers partly in response to increased competition.

The return to shareholders of the major banks peaked in the early 1980s, before the introduction of the major deregulation of the 1980s (Figure 14). The return on shareholders' funds averaged 16 per cent in the early 1980s before falling to an average of 14 per cent in the second half of the 1980s. Returns fell sharply in the early 1990s as a result of the slowdown in economic activity, the collapse of asset prices and the consequent rise in bad debts and non-performing loans. While returns were well in excess of the yield on Government bonds before the 1980s, the gap narrowed substantially during the 1980s. Similarly, returns in banking were high relative to other industries in the early 1980s but declined relatively over the course of the decade (Figure 15).

Thus, it appears that returns to banks' shareholders declined in an absolute and relative sense over the 1980s despite the rapid growth in the economy and bank balance sheets that occurred for most of this period.^[30] This is consistent with banks doing a higher volume of activity at lower margins, on average over the whole period, and suggests that deregulation and increased competition could have reduced bank profit margins and the return to bank shareholders somewhat. It is, however, difficult to assess whether there has been a structural decline in the profitability of banks. Most of the fall at the end of the 1980s occurred because of a rise in bad debts and non-performing loans which, in turn, reflected the process of deregulation as banks sought higher yielding more risky loans in the face of rising deposit costs. As the effects of these problems are worked out, bank profitability could recover.

3.3 Summary

Several points are worth noting from the previous section:

• there has been a rapid expansion of the financial system along with household and corporate sector balance sheets;
• the corporate sector became more reliant on debt in some countries (particularly those that started with lower leverage). Innovation and globalisation gave some firms greater access to securities markets as a source of funds;
• household financial assets grew rapidly, with much of the growth being in assets held in pension funds, equity and unit trusts. Nonetheless, deposits at banks remain an important part of the households asset portfolio. The household sector also became more indebted in some countries; and
• banks shared in this financial expansion to a greater or lesser extent in most countries. In almost all countries, banks extended the range of services they provide, and non-interest income became a more important source of revenue. Lending to the corporate sector remained an important part of their assets. In many countries, banks funded a good part of their expansion by borrowing from domestic and international securities markets. Banking profitability appears to have fallen somewhat.

Footnotes

Data definitions and sources for all graphs and tables can be found in the Data Appendix. [9]

Differences in accounting practices across countries make it difficult to compare gearing ratios. Adjusting for these differences would still leave Japan, Germany and France with high leverage. O'Brien and Browne (1992). [10]

Borio (1990) finds that differences in the tax system and structural impediments to the development of stock markets can account for some of the difference. He argues that the bulk of the difference is due to how information problems are resolved between borrowers and lenders. [11]

Bisignano (1990) and Borio (1990). [12]

Frankel and Montgomery (1991). [13]

Bisignano (1990, pp. 3–4). [14]

For an analysis of these developments in Japan see Bank of Japan (1991). [15]

For example see Edwards (1993). [16]

Gertler and Hubbard (1988). [17]

O'Brien and Browne (1992). [18]

Boyd and Gertler (1993) discuss US developments. [19]

This figure records bills held on bank balance sheets. It, therefore, understates banks' involvement in providing bill finance. Banks facilitate the issuance of a proportion of the Bills figures quoted in Table 3 by accepting or endorsing them. [20]

For a discussion of the behaviour of household balance sheets in the 1980s see Callen (1991). [21]

Edey, Foster and Macfarlane (1992) discuss these issues in more detail. [22]

The numbers in Graph 8 are from a sample of 96 companies listed on the Stock Exchange. They are not directly comparable to the ABS figures underlying Table 3. [23]

See Lowe and Shuetrim (1992) and Mills, Morling and Tease (1993) for a discussion of corporate balance sheet behaviour. [24]

See Grenville (1991) for a discussion of the constraints on financial activity in the 1970s. [25]

Grenville (1991) analyses the effects of deregulation. [26]

The deposit share of banks fell from nearly 80 per cent in 1968 to 55 per cent in 1982 and has since recovered strongly. See Battellino and McMillan (1989). [27]

Under the SRD banks were obliged to hold a proportion of their Australian dollar deposits in accounts paying below-market interest rates at the Reserve Bank. [28]

The credit equivalent basis measures off-balance sheet business on the basis of the credit exposure associated with the various transactions. [29]

Ackland and Harper (1990) also provide evidence of a deterioration in balance-sheet measures of bank performance. On the other hand, analysis based on the risk-adjusted returns available to bank shareholders does not support a conclusion that bank shareholders earned excess returns prior to deregulation or that they experienced a fall in their returns subsequently (Harper and Scheit (1991)). [30]