Accessibility and Bank Mergers
in British Columbia

A Report Prepared for the British Columbia Task Force on Bank Mergers
Professor Paul Bowles
Department of Economics
University of Northern British Columbia
3333 University Way
Prince George
B.C. V2N 4Z9
28 July 1998


A fully `accessible' financial system would be one in which all individuals would make their decisions in an environment characterised by a competitive financial system. Such accessibility has two dimensions which arise directly from the functions of the financial system. These functions include the collection, administration and aggregation of savings and the distribution and supervision of lending to borrowers. The two dimensions of accessibility derived from this are access to financial services as a depositor and access to capital as a borrower.

The first dimension, access to financial services, involves both a geographic component and in income component. The geographic component refers to the degree to which access to a competitive banking system is available to residents of particular geographic locations such as those in rural and northern British Columbia. The income component which refers to the extent to which people on low incomes have access to financial institutions.

As a background to the proposed mergers it should be noted that, from a geographic accessibility perspective, a number of communities are already faced with limited competition in the provision of financial services. In particular, it should be noted that 119 communities out of the 172 communities with less than 100,000 inhabitants considered here are served by 3 or fewer financial institutions. As a result of the mergers, the percentage of communities-with-at-least-one-merger-bank in this `high concentration' category will increase from 53 per cent to 63 per cent of communities. If we consider the entire sample (i.e. and include communities not presently served by a merger bank), the percentage of communities in the `high concentration' category increases from 69 per cent to 75 per cent as a result of the mergers.

A `Branch Closure Vulnerability Index' is constructed to assess how vulnerable communities are to branch closures as a result of the mergers. 75 per cent of the communities for which the Index was computed were either in the `moderate' or `high' vulnerability categories. The results indicate therefore that the bank mergers are estimated to have a significant effect on the number of communities that can be considered `high concentration' communities and that a significant number of communities are moderately to highly vulnerable to a post-merger bank branch closure.

With respect to access to financial services by low income individuals, barriers to accessibility are found to exist.

The second dimension of accessibility, access to capital, refers to the fact that particular groups in society, who may or not be geographically concentrated, often experience difficulties with access to financial capital. This has both equity and efficiency implications. The groups often included as experiencing such difficulties include small businesses, low income individuals, and aboriginal groups.

Small businesses report significant degrees of dissatisfaction with the current operations of banks. This dissatisfaction relates to both the availability of credit and the terms on which it is offered. It relates to the turnover of account managers at banks which reduces the monitoring abilities of banks and hence increases the problems faced by small businesses in dealing with their banks. The effects of the proposed bank mergers on small businesses are assessed and it is concluded that the effects will be negative.

Lending to low income individuals and aboriginal groups are examined. These markets are currently not well served by the chartered banks and the impact of the proposed mergers is therefore limited.

This Report concludes that the proposed mergers will have a negative effect on access to financial services and on access to capital in British Columbia.

A number of responses to the current situation are discussed. These responses may come from the banks themselves, from the market through competitive behaviour and reduced barriers to entry and from governments in the form of policy and regulatory responses. Each of these components is analysed.

Accessibility and Bank Mergers in British Columbia [1]

I. Introduction

The financial system plays a critical role in the economy; it is responsible for the collection and aggregation of savings and for the distribution and supervision of lending to borrowers. It is therefore a critical intermediary between savers and lenders which affects the performance of the `real economy'. Since modern economies are monetary economies, the operations of the financial system also influence how individuals are able to participate in the economy, whether as consumers, savers, entrepreneurs or investors.

In British Columbia, in 1997, the main type of financial institution, the chartered banks had a network of over 900 bank branches, had personal savings deposits of $42 billion, had issued credit to over 92,000 small and medium business, and had outstanding residential mortgages of over $41 billion. [2] Obviously, any change in the operations to the banking system is a subject requiring thorough study. In this paper, the impact of the proposed mergers between the Royal Bank (RB) and the Bank of Montreal (BMO) and the Canadian Imperial Bank of Commerce (CIBC) and the Toronto Dominion (TD) are analysed for their impact on `accessibility'.

In the next section, `accessibility' is defined and analysed. In section III some possible policy responses are identified.

II. Accessibility

A fully `accessible' financial system would be one in which all individuals would make their decisions in an environment characterised by a competitive financial system. Such accessibility has two dimensions which arise directly from the functions of the financial system. These functions, as we have seen, include the collection, administration and aggregation of savings and the distribution and supervision of lending to borrowers. The two dimensions of accessibility derived from this are access to financial services as a depositor and access to capital as a borrower. Of course, typically individuals and businesses are both depositors and borrowers but since the issues of accessibility are conceptually different we separate them here for analytical purposes.

The first dimension, access to financial services, involves both a geographic component and in income component. The geographic component refers to the degree to which access to a competitive banking system is available to residents of particular geographic locations, especially those outside of the major conurbations, such as those in rural and northern British Columbia. The issue which needs to be addressed therefore is how individuals living in low population density areas are likely to be affected by the proposed bank mergers. The income component which refers to the extent to which people on low incomes have access to financial institutions. How this will be affected by the proposed mergers needs to be analysed.

The second dimension, access to capital, refers to the fact that particular groups in society, who may or not be geographically concentrated, often experience difficulties with access to financial capital. This has both equity and efficiency implications. The groups often included as experiencing such difficulties include small businesses, low income individuals, and aboriginal groups.

The analysis presented here of accessibility to the `financial system' is restricted to all deposit-taking institutions in B.C., namely, the chartered banks, the trust companies and the credit unions. Other financial services - such as stockbroking and mutual funds - are also provided through a variety of other financial intermediaries. For consumers, these additional intermediaries provide important choices. However, these services lie beyond the scope of this paper and analysis is restricted to the financial system as defined above. In addition, the paper considers access to capital by entrepreneurs and firms; access to capital by individuals in the form of mortgages, for example, is excluded from this study.

The proposed bank mergers pose problems for theoretical and empirical analysis. Theoretically, we need to have a good grasp of the relationships between concentration and competition. In the analysis that follows, data will be provided on the presence of financial institutions in B.C.; the assumption being that individuals living in communities with few financial institutions have limited access to a competitive financial system. Such an assumption is consistent with the standard theoretical approach to industrial organisation which focuses on measures of industry concentration as an indicator of industry structure and a predictor of industry conduct.[3] However, this approach has recently been challenged by theoretical work which focuses on the `contestability' of markets rather than on actual competition. Contestability in this sense refers to the ease of access by new market entrants and hence the discipline that this imposes on existing firms. If this is the case, then high concentration ratios do not necessarily imply uncompetitive markets. The Royal Bank (1997: p. 18) makes this point, noting that "an industry with a small number of firms can be very competitive if the market is "contestable", that is, competitive discipline is brought about through the threat of new entrants."

Evidence on contestability does, however, vary. For the U.S., for example, Amel and Liang (1998) find that potential competition is not as important as actual competition. Thus, new entrants tend to reduce profits (suggesting that existing firms are making rents) and they conclude, as Rhoades (1998: p.6) notes that "bank markets are not contestable." Studies by Frame and Kamershen (1998), Jackson (1998), and Sharpe (1998) also support the view that banks in the U.S. exercise significant market power and that in areas where banks have higher concentration ratios this results in less competitive practices being observed. However, Shaffer's (1989) study finds that collusion between banks is rejected but that the hypothesis that the U.S. banking industry is characterised by perfect competition cannot be rejected. Since the top 5 banks' share of total domestic financial assets in the U.S. is less than 10 per cent whereas it is closer to 60 per cent in Canada, the finding by some studies that market power exists in the U.S. banking sector might be expected to lead to empirical results indicating that market power exists in the Canadian banking industry. [4] However, studies by Shaffer (1993) and Nathan and Neave (1989), for example, do not support this conclusion. Studies in Europe also cast doubt on the relationship between industry concentration and industry conduct. [5]

While accepting that the evidence on the contestability of bank markets is not conclusive, this paper nevertheless takes the position that, as a first approximation, accessibility to a competitive financial system can be gauged by the number of financial institutions in communities. This requires the additional assumption that the relevant space over which `access' should be considered is the `community'. Or, in other words, that the `market' for financial services is a local one. In support of such a position, the Federal Competition Bureau notes that (1998: p. 12) "consumers of certain types of banking services may be unwilling to switch to suppliers outside of their local areas in response to an increase in the prices of these services in their own areas. That is, for certain types of products, geographic markets may be local, comprising only a limited geographic area." This would reasonably seem to be the case for retail banking facilities. [6] A focus on the rural retail banking markets supports the approach taken in this paper of analysing concentration levels since the `contestability' approach has not systematically analysed these markets.

Empirically, the challenge is that the proposed mergers have not taken place and there have only been a limited number of major mergers in the Canadian financial system in recent years, the most prominent being mergers between chartered banks and trust companies. Thus, there is little previous data with which can be used to make reliable predictions about the impact of the proposed mergers. Bank mergers have, however, occurred elsewhere most noticeably in the U.S. There were over 6,300 bank mergers in the U.S. between 1980 and 1994 and more have been announced since partly as a result of the passage of the Branch Banking and Efficiency Act in 1994 which allowed inter-state banking by bank holding companies in 1995 and inter-state branch banking in 1997. [7] Even though the context for bank mergers is different in the U.S. in that mergers there are creating a national banking system whereas such a national system already exists in Canada, nevertheless many of the issues are the same and U.S. studies, as well as other comparative studies, are therefore cited here where appropriate.

The discussions of access to financial services and access to capital are organised as follows. In each case, an analysis of the present situation is provided. This is followed by an analysis of the impact of the proposed bank mergers on the current situation and, in particular, on whether accessibility is likely to be enhanced, remain unchanged or decreased. That is, the mergers are analysed in terms of whether they are likely to increase "accessibility" or contribute to "financial exclusion". Arising out of this analysis some responses and regulatory options are discussed in Section III.

II.a Access to Financial Services

II.a.i. Geographical Access to Financial Services

Concerns about the accessibility of financial serves as a result of bank mergers is primarily an issue of concern to small and low population density communities. It is generally accepted that competition between financial institutions is lower in rural areas than in the major centres. For example, data for the U.S. show that the standard measure of industry concentration, the Herfindahl-Hirschmann Index (HHI) was 2.3 times higher for banking organisation in non-metropolitan statistical areas (non-MSAs) than in MSAs. [8] Furthermore, almost all non-MSAs had an HHI exceeding the high concentration threshold of the Department of Justice-Federal Trade Commission horizontal merger guidelines. [9] In Canada, data on assets per bank branch are not available and so such clear indicators of the differences between urban and rural concentration ratios in the banking system cannot be used for analytical purposes. As an entry point into this discussion, therefore, it has been necessary to compile geographic inventory of the existing distribution of financial institutions in B.C. to see, as a first cut, the availability of financial institutions in small town B.C. A liberal definition of "small town" has been used to include communities of up to 100,000 inhabitants. Only the major conurbations have therefore been excluded.

Appendix Table 1 provides an overview of current geographic accessibility to deposit taking institutions in B.C. Data for this Table was complied from information on the location of financial institutions' branches in B.C. provided by the Canadian Payments Association. To this list of communities, communities classified by Statistics Canada Population Census data as villages, towns, district municipalities and cities was added. The list of communities in Appendix Table 1 is based on the combination of these two data sources. [10] As can be seen in Appendix Table 1, a significant number of small towns and cities in B.C. already have very limited access to the banking system; 23 communities have no financial institutions at all, 55 have no chartered bank, and 52 communities are served by only 1 financial institution.

As a background to the proposed mergers, therefore, it should be noted that a number of communities are already faced with limited competition in the provision of financial services. In particular, it should be noted that 119 communities out of the 172 communities considered here are served by 3 or fewer financial institutions. The Competition Bureau guidelines, recently reaffirmed as applicable to the financial sector, state that market shares of less than 35 per cent do not generally give to rise to concerns about competition in any particular market. Individual firm market shares in excess of 35 per cent may, however, give rise to such concerns and warrant investigation by the Competition Bureau. On the assumption that financial institutions all have an equal market share in small communities, communities with three or less financial institutions may be thought of as `high concentration' communities. [11] This is summarised below in Table 1.

Click here to view Table 1.

The findings reported in Table 1 indicate that 69 per cent of B.C.'s communities with less than 100,000 population are `high concentration' communities. As might be expected, communities with smaller populations have, on average, fewer financial institutions.

The proposed mergers will undoubtedly intensify this high concentration for some communities and extend it to others. Can we say how serious this is likely to be? If the proposed RB-BMO and CIBC-TD are approved, this will have implications for the number of banks operating in communities and also, in all probability, the number of bank branches. As a first look at this, Appendix Table 2 shows the percentage reduction in the number of banks in the 112 communities served by at least one merger bank at present as a direct consequence of the mergers.

The number of banks in any community could be decreased by a maximum of 2 as the result of mergers. The implications of this for the number of communities which will fall into the `high concentration' category is shown below in Table 2.

Click here to view Table 2.

Table 2 indicates that the mergers will increase from 59 to 71 the number of communities-with-at- least-one-merger-bank in the `high concentration' category. Put alternatively, the percentage of communities-with-at-least-one-merger-bank in the `high concentration' category will increase from 53 per cent to 63 per cent as a direct result of the mergers. If we consider the entire sample (i.e. and include communities not presently served by a merger bank), the percentage of communities in the `high concentration' category increases from 69 per cent to 75 per cent as a result of the mergers.

It should be noted that since the merger banks are typically bigger, on average, than other financial institution the reduction in the number of financial institutions in a community will underestimate the increase in concentration of the industry in that location. On the other hand, the post-merger reduction takes no account of any potential new entrants into these markets which might occur as a competitive response to the mergers and the reduction in the number of financial institutions and will therefore overestimate the reduction in competition as a result. The data in Table 2 should be read in this light. Even so, the data do indicate that a number of B.C. communities do face the prospect of a significant percentage reduction in the number of banks serving their communities if the mergers take place.

There are also likely to be reductions in the number of bank branches if the two bank mergers are approved leading to a reduction in access to banking institutions. While the precise number of bank branch closures may be open to dispute, the fact is that there will be closures. [12] This presents something of a paradox in that most empirical work on mergers in the U.S. indicate that bank mergers have not increased cost efficiency i.e. have not reduced costs as a ratio of bank assets or bank revenues. [13] Furthermore, in the U.S. at least, most bank merger announcements have not increased the share price of the merging firms indicating that the stock market's assessment of bank mergers is also that efficiency is not increased. However, although efficiency has not been shown to increase, cost cutting measures including staff reductions and branch closures have been prominent parts of merged banks activities.

As Rhoades (1998: p. 275) explains: "In studying the possible gains from mergers, it is important to distinguish between cost reductions and efficiency improvements; they are not synonymous. Reductions in operating expenses may result from cutting employees, closing branches, consolidating headquarters offices, closing computer and back-office operations, and so forth. Such reductions in expenses, however, do not automatically translate into improvements in efficiency as by an expense ratio, such as expenses to assets or revenues. Reductions in expenses may be accompanied by corresponding reductions in assets and revenues, which simply represent shrinkage of the firm rather than efficiency improvements. An improvement in efficiency requires that costs be reduced be reduced by more than any decline in assets (revenues). Failure to distinguish between cost reduction and efficiency gains may at least partly explain the difference in views between bankers, who often emphasize the cost reductions to be achieved from mergers, and researchers, who generally study the efficiency effects of mergers." (Emphasis in original)

Thus, bank closures will occur as a result of the proposed mergers and should be studied separately from the efficiency effects of mergers; branch closures neither imply nor are implied by increased efficiency in an economic sense.
Furthermore, the merger banks themselves have noted the excess capacity in the Canadian banking system. As the Royal Bank (1997: p.8) note: "one indication of [the] technology-driven rise in capacity is the recent accelerating increase in real assets per bank branch in Canada. Given that the financial services industry is a relatively mature one in Canada and other developed countries, increasing capacity implies over-banked markets and suggests substantial consolidation will be required." While this does not speak directly to the issue of bank branch closures it is certainly implied by the recognition of "over-banked markets".

The time period over which bank branch closures might occur depends on the banks themselves. The RB and BMO have publicly declared that no town in which they are only bank will loose that bank for a period of 5 years. For other communities, evidence from the U.S. indicates that most of the cost savings from bank mergers were achieved within one year and practically all savings made within three years. (See Rhoades 1998). The one unexpectedly high cost identified in bank mergers was the costs associated with merging data systems. Given the problems associated with the "year 2000" phenomenon this may act as a brake on the speed of merger induced cost savings at this historical juncture.

Will possible bank branch closures be a concern in B.C.? For any particular community, if the likelihood of bank branch closure is small and if there are a significant number of other deposit taking institutions also operating in the community then that community may be deemed to have only a low vulnerability to such a closure. However, if there is a strong likelihood that a community will lose a bank branch as a result of the mergers and there are few or no other financial institutions, then that community may be regarded as being highly vulnerable to a branch closure. To consider this in a more sophisticated way, we therefore construct a Branch Closure Vulnerability Index (BCVI) which indicates how vulnerable particular communities are to bank closures. It should be noted that the index does not measure the vulnerability of any particular branch bank to closure; rather it measures the vulnerability of particular communities to bank branch closures.

As noted above, a community's vulnerability to bank branch closures depends upon two factors: the likelihood that a bank branch will close as a result of the mergers and the impact of such a closure on the degree of financial competition within the community. The BCVI attempts therefore to measure both the likelihood of bank branch closure and the seriousness of that closure for the competitive environment in that community. These are two separate components of the BCVI in that some communities may face branch closures but there may be a large number of alternative banking institutions while, on the hand, some communities may few alternatives but may not be at risk of losing a bank branch.

Construction of any index is fraught with methodological difficulties. [14] The choice of indicators to include in an index, their measurement and their aggregation are all matters for debate. [15] Many indices do, nevertheless, become widely accepted by researchers and policy-makers. The BCVI is not one of these; indeed, I know of no previous attempt to construct one. Thus, the index should be used with caution and interpreted as a useful for device for framing the impact of bank mergers rather than as a mechanical tool with which to make definitive predictions. Much depends on the actions of the banks themselves.

The Index is constructed as follows. For each community with at least one merger bank branch, the following variables were calculated as indicators of the likelihood of bank branch closures:

(a) branch density. The rationale behind this variable is that a merger bank branch is more liable to closure the smaller the number of clients that it serves. [16] Data on number of clients per branch or on branch profitability are not available. As a proxy measure, population per financial institution branch is used; [17]

(b) merger bank duplication. The rationale for this variable is that the higher the number of merger bank branches there are in a particular community the greater the probability of a post-merger closure. The number of merger bank pairings was used where a pairing indicates the presence of a branch from both the CIBC and TD or from both the RB and BMO;

(c) relative income. Communities with lower per capita incomes may be considered to have lower demand for financial services (especially the most profitable investment-related services) and may be more liable to bank closure. [18] Average community income as a percentage of average Provincial income was therefore calculated.

In addition, the following variables were calculated as indicators of the impact of a branch closure on the community:

(d) merger bank dependence. The impact of a branch closure will depend on the number on merger bank branches that it currently has. If a community has only one merger bank branch that is closed as a result of post-merger rationalisation then this will be have a more serious impact on the community than if there are several merger banks. The number of merger bank branches was therefore calculated;

(e) competitive environment. The reduction in competition which will arise from a merger bank branch closure will depend on the number other financial institutions. Thus we include the number of other non-merger chartered banks, the number of trust companies, and the number of credit unions serving the community. The higher the number of other financial institutions' branches the less vulnerable the community to bank mergers. Ideally, this could be adjusted by asset size of institution to indicate how large these other institutions' are; however, asset figures per financial institution are unavailable and so a simple numerical measure has been used.

Aggregation. In constructing an index of any kind, weights must be attached to the variables comprising the index. These weights are typically assigned by the researcher based on a priori reasoning or based on results of empirical studies. In some cases, a principal components approach is taken, which assigns weights based on the information content of the data itself. In the absence of extensive empirical work in this area, a very simple procedure has been adopted in constructing the BCVI. Each variable is assigned a value between zero and three based on whether it meets pre-specified critical values or not. The critical values chosen by the researcher are shown below in Table 3.

Click here to view Table 3.

These critical values were based on judgements made by the researcher. A sensitivity analysis has not been undertaken to examine the effects of changing these critical values. As such, additional caution should be used in interpreting the Index.

Based upon this, scores were aggregated using equal weights to produce the BCVI ranging from 2 to 12. Appendix Table 3 provides the full results for the 92 communities for which all of the data required to construct the Index was available. It should be noted that data unavailability was an issue for 20 of the smallest communities; the sample will therefore underestimate the vulnerability of communities since those excluded are likely to be biased towards the high vulnerability end of the Index. A summary of the results is given in Table 4 below. Communities with BCVI scores of 2-5 are defined as `low vulnerability' communities, 6-8 are defined as `moderate vulnerability' communities and 9-12 are defined as `high vulnerability' communities.

Click here to view Table 4.

Table 4 indicates that 75 per cent of the communities for which the Index was computed were either in the `moderate' or `high' vulnerability category.

The preceding analysis has pointed to the effects that bank mergers may have the geographic accessibility of financial services. The results indicate that the bank mergers are estimated to have a significant effect on the number of communities that can be considered `high concentration' communities and that a significant number of communities are moderately to highly vulnerable to a post-merger bank branch closure.

It should be noted, however, that these estimates should be considered as maximum estimates since, as pointed out above, no account has been taken of competitor responses. These will be considered further in section III. Here, it is useful to note that there are alternatives to branch banking. That is, financial services may be delivered by platforms other than branches of financial institutions. Two such alternative delivery systems are considered briefly here, namely, Automated Banking Machines (ABMs) and on-line banking.

In terms of the availability of ABMs, according to Statistics Canada (1998), "with 18,426 ABMs, about six for every 10,000 people, Canada was second only to Japan in making the machines available to customers." With respect to the Internet, use by Small and Medium Enterprises (SMEs) has increased in rapidly in Canada. By March 1998, 43 per cent of business owners surveyed by the Canadian Federation of Independent Business (CFIB) said they had access to the Internet. This is expected to rise to between 50 and 60 per cent by 1999. The figures for B.C. are very close to the national figures.

Alternatives to branch banking do, therefore, exist. The question is the extent to which they are a perfect substitute for branches of financial institutions. In this respect, a report by Informetrica (1998: p.5) argues that "many small businesses as well as older customers depend on the personalized services to which they are accustomed." The CFIB (1998b) cautions that "for now, in addition to experimenting with new electronic service methods, governments, financial institutions and other major suppliers to SMEs will still have to rely on traditional delivery mechanisms to deal with this market place."

These concerns are supported by studies from the U.S. Rhoades (1998: p.4), for example, writes that in the U.S., the number of ABMs "increased in number from 18,500 in 1980 to 109,000 in 1994. Despite these large numbers, ATMs have proven to be primarily a convenient cash dispenser rather than anything approaching a general substitute for the traditional banking office... Personal computers, while receiving a great deal of publicity as a tool for electronic banking, have thus far been used very little by bank customers in utilizing banking services, and some analysts doubt that the PC will ever be a substitute for a branch".

Thus, while ABM availability in Canada is already high by comparative standards and while on-line banking is also likely to prove increasingly useful for SMEs and more affluent households, these delivery systems cannot be viewed as perfect substitutes for physical branches.

II.a.ii Low Income Access to Financial Services

It is estimated that 8 per cent of consumers in Canada with an annual income of less than $25,000 do not have bank account; this translates into at least 400,000 individuals. [19] For some this may be a matter for a choice but for others is undoubtedly an issue of `financial exclusion'. Access to banking facilities is taken for granted for the vast majority of Canadian residents. However, for low income individuals, no such presumption can be made. Being unable to participate in a the `monetary mainstream' imposes costs on low income individuals since they are forced to use more expensive financial institutions such as Money Marts which have increased significantly in number in Canada during the past decade and constitutes a form of social exclusion. This form of exclusion is likely to become worse over time as trends in income inequality and poverty rates have deteriorated during over the past 15-20 years in B.C. For example, average real income for households in B.C. fell by 7 per cent between 1981 and 1994. Furthermore, the degree of inequality around this declining mean increased. [20]

The chartered banks have responded to criticisms related to accessibility by low income individuals by establishing a voluntary code of conduct, set out by the Canadian Bankers Association, which is intended to ensure access to basic banking services by all Canadians. The code revises the identification requirements for opening a bank account and removes the employability criteria. Critics note that implementation is a problem and that credit checking requirements are too stringent. [21]

The problem of low income individuals' access to the financial system is exacerbated when, as if often the case, low income individuals are geographically concentrated. This has happened in many U.S. cities where neighbourhoods have been subjected to `financial abandonment' as financial institutions have closed branches in low income neighbourhoods. Similar trends have been documented in the U.K. as well. In B.C., the most obvious case of this is found in the Downtown Eastside of Vancouver. [22] After the Bank of Montreal closed its branch in the area in the mid-1990s none of the major financial institutions had a presence in the area. The BMO had previously worked with the Downtown Eastside Residents Association (DERA) and the Ministry of Social Services in providing financial services to residents as part of the Downtown Deposit Project. Following the BM's closure, however, residents had to utilise banks in other adjacent neighbourhoods. However, while these other banks were within easy geographic distance there were a number of `invisible barriers' to low income individuals such as the banks' hiring of private security guards. A Provincially government supported financial institution, the Four Corners bank, has now opened in the Downtown Eastside area. The Downtown Eastside represents the most extreme example of financial abandonment in urban B.C.; the impact of the bank mergers is of little relevance here.

The impact of the mergers on the accessibility of low income individuals in general depends upon the restructuring plans adopted by the post-merger banks. In proposing their mergers, the banks hope to increase efficiency. As indicated above, this has proven problematic to achieve in practice. However, a study of bank mergers in the U.S. by Akhaven, Berger and Humphrey (1997) utilises a broader concept of efficiency than that typically employed in econometric studies which leads them to conclude that merged banks may increase "profit efficiency" through reallocating resources within the firm and shifting the bank's "output" to products with higher revenue. If mergers led to a shift in banks' output in this way (pressures which are presumably present even in the absence of mergers) which activities would be likely to expand?

Statistics Canada (1998) reports that "in 1996, the profit margin ratio [for all deposit-accepting institutions] was 29 cents per $1 of all services produced ... With more than 40 cents per $1 of services produced, electronic financial services, treasury and investment banking and corporate institutional finance were the activities which posted the highest profit margin ratio in 1996 for all institutions." It is these areas of services, therefore, that we would expect the banks to focus on expanding. However, these financial services are typically not well utilised by low income individuals with the probable exception of ABM electronic services. [23] Retail banking, access to which by low income individuals has been at issue, has the lowest profit margin.

II.b Access to Capital

In section II it was suggested that small businesses, low income individuals and aboriginal people's are typically identified as facing greater barriers to the access of capital than other borrowers. Examples of why these groups may experience difficulties include the following. Small businesses may pose challenges to lending institutions since they may have limited collateral, be operating in markets with low profit margins and fluctuating demand, and where the success of the business may depend to a large degree on the abilities and character of the owner. Aboriginal people's title to land has been a contentious issue and without clear title, collateral requirements cannot be met. Low income individuals also face similar problems meeting traditional collateral requirements. Each of these groups in considered in more detail below.

II.b.i Small Businesses

When making lending decisions banks face monitoring problems. They have incomplete information of borrowers and must use various mechanisms to assess the performance of potential borrowers while keeping risk at acceptable levels. Riding et al (1994: p.7) argue that "the decision to grant credit to a business applicant ... incorporates objective information and criteria as well as subjective and qualitative impressions. These criteria are frequently referred to as `the 5 C's of commercial lending': Capital, Collateral, Conditions, Capacity and Character." These criteria are typically easier to apply when dealing with large corporate customers, with long histories, high collateral, stable and predictable financial operating requirements and a competent senior management team. The costs of administering large loans are also relatively low. Small business (as well as new firms), on the other hand, pose more of a challenge to banks for some of the reasons outlined above and it is not surprising therefore that the bank-small business relationship in Canada has not always been a smooth one.

Evidence of this is provided by the Canadian Federation of Independent Business (CFIB). The CFIB member firms tend to be more established and to be slightly larger than the average small business in Canada. As a result, there may be a bias which would result in these firms having less problems with financial institutions than other small businesses.

CFIB data indicate that 75 per cent of SMEs rely on financial institutions for business credit. [24] Credit from financial institutions therefore plays a key role in the development of SMEs. However, while the majority of SMEs are satisfied with their financial institutions, significant numbers report concerns over the availability of funding to their firms. This level of concern has increased during the past decade as indicted below in Figure 1:

Click here to view Graph
Use browser "Back" button to return to report

Source: CFIB (1997: p. 12)

CFIB (1998d) survey data indicate that access to bank credit was ranked as the seventh most important factor to increasing the employment plans of small business in B.C. in 1998. A significant concern for SME owners in the pre-merger context is therefore the availability of credit. [25] Data provided by the Canadian Bankers Association confirms the continued relevance of this concern. Loans to SMEs have fallen from 26.86 per cent of the major banks' total lending at the end of 1995 to 25.04 per cent of total loans in the third quarter of 1997, the lowest level since reporting started in 1995. [26]

Another consistent concern voiced by the small business community has been with the level of account manager turnover. As discussed above, banks face problems with monitoring small enterprises. Personal knowledge of the firm and the firm's owner and/or manager is therefore an important resource for bankers when making loan decisions. It is also a resource which is valued by small business owners. Bankers with less knowledge and less familiarity with the small business are likely to find it harder to fulfill this monitoring function. In British Columbia, 28.2 per cent of firms reported that they had 3 or more account managers in the previous three years. [27] We would expect to find that the probability of loan application rejection would rise with the turnover rate of account managers at the banks. This is supported by the survey evidence produced by the CFIB reported below in Figure 2.

Click here to view Graph
Use browser "Back" button to return to report

Source: CFIB (1997: p. 11)

Furthermore, account manager turnover for small businesses appears to have changed little in the past ten years as shown in Figure 3 below.

Click here to view Graph
Use browser "Back" button to return to report

Source: CFIB (1997: p. 11)

As indicated above, bank's relations with small business are not purely mechanical exercises but rely on the judgement and knowledge of account managers with banks. This means that loan officers must develop their own rules of thumb. Riding et al (1994: p.7), quoting a study by Jankowicz and Hisrich, note that "among other findings, [Jankowicz and Hisrich] conclude that loan officers prefer `stability and conformist behaviour'." Individuals who do not fit this pattern may therefore find credit financing additionally difficult to obtain. The CFIB report (1995: p.1) that "discrimination against women small business owners by Canada's major financial institutions is widespread across the country ... The discrimination is twofold: women seeking finance are refused 20 per cent more often more often than men; and women are regularly charged a higher rate of interest than men. These findings take into account differences in the size, age and type of business operated by men and women."

In summary, SMEs report significant degrees of dissatisfaction with the operations of financial institutions. This dissatisfaction relates to both the availability of credit and the terms on which it is offered. It relates to the turnover of account managers at banks which reduces the monitoring abilities of banks and hence increases the problems faced by small businesses in dealing with their banks. There is also evidence that women business owners face greater problems than their male counterparts. Summarising previous research, Riding et al (1994: p.11) argue that "first, non-rate aspects of the relationship between small businesses and banks seem to be valued by business owners. Second, banks' performance on those aspects of the relationship that seem to be valued the most are not, in the aggregate, rated very highly by small business owners."

How would bank mergers change this? The affect of bank mergers on small businesses is a complex and controversial topic. The issue has been much debated and investigated in the U.S. where concerns have been expressed about the potential adverse impact of bank mergers on the volume of lending to small businesses and farms. The evidence from empirical studies (for example, Peek and Rosengren, 1998; Keeton 1996; Jayaratne and Wolken, n.d., and Cole and Walraven, n.d.) is mixed. The regional structure of banking makes the comparative experience less relevant in this respect to an analysis of the Canadian case. Here, we highlight those factors which have been identified as important in the debate on small business financing and predict in which direction they might be expected to change given the specific conditions of British Columbia.

Consider first the availability of credit:

Bank mergers may increase lending to small firms and rural communities if:

(a) The merged bank is less prone to risk from a local economic downturn;
(b) The merged bank is able to provide more liquidity;
(c) The merged bank may be better able to meet loan surges in one location by reallocations within the organization.

Bank mergers may decrease lending to small firms and rural communities if:

(d) Loan centralization decisions reduces knowledge of local area and businesses;
(e) The merged bank may have better uses for the funds.
(f) Larger institutions are more risk adverse and reduce lending to relatively risky sectors such as the small business sector;

Consider next the conditions of credit and quality of service:
Possible positive effects of mergers:

(g) The merged bank will now be able to employ specialist small business loan managers;
(h) The merged bank will increase efficiency and lower service charges.

Possible negative effects:

(i) The merged bank will result in less competition in rural areas and service will deteriorate and/or service charges increase;
(j) The merger will lead to a rationalisation of staff responsibilities and increase loan account manager turnover;
(k) Bank mergers will reduce competition and reduce the possibilities for small business owners to switch banks.

If we consider each of these effects for British Columbia, then it is important to note that the proposed mergers are between banks which are already large in size and national in scope. The possible advantages which might result from a merger from a small, regional bank and a larger bank in terms of providing greater liquidity and ability to ride out local economic downturns are therefore negligible. Thus, variables (a), (b), (c) and (e) are likely to be zero in the B.C. case. The size of the existing banks means that specialist loans officers are already in place and that further gains through this channel (variable (g) above) are also likely to be small.

Variable (d) is likely to be a major concern to small businesses. As argued above, loan monitoring for small businesses is best done by account managers with detailed knowledge of the firm and based on personal knowledge of the loan applicant. Since 1981 the Canadian banks have, at various speeds, implemented loan centralization systems. However, this has been followed by periods of decentralisation and then recentralisation. At present, the degree of loan centralisation various considerably by bank. For example, in the Northern B.C. and Yukon regional area, loan approval decisions which can be made at the regional office in Prince George varies from $250,000 to $5 million. Thus, banks have different marketing strategies and organisational cultures which lead to different degrees of loan centralisation. A post-merger loan approval process would therefore depend on the strategy and organizational culture adopted by the post-merger bank. This is, of course, difficult to predict although it might be thought that a post-merger bank intent on a global corporate strategy would likely be more hierarchical in its organisation. If this were indeed the case, then lending to small business is likely to suffer. [28]

With respect to variable (f), studies from the U.S. indicate that larger financial institutions are more risk averse and hence lend less to risky sectors. [29] In Canada, a CFIB survey (1998c: p.2) of its membership found that, with respect to satisfaction with financial institutions' willingness to lend to SMEs, "the large financial institutions do not perform as well as the smaller ones. Although Royal Bank and CIBC are bank leaders on the SME market, these institutions find themselves at the very bottom in terms of banks' willingness to lend."
With respect to variable (h), evidence on mergers and efficiency does not indicate that mergers are unambiguously more efficient and the relationship between bank size and service charges also shows no negative relationship.

Variable (i), representing service charges levels, is also of concern to small businesses in B.C. As indicated above in Section II.a.i. A significant number of small communities are already high concentration communities and this would be increased by the mergers. CFIB surveys indicate that average service charges increases are a function of the degrees of competition in a community. See Appendix Figure 1.

With respect to variable (j), bank mergers will inevitably lead to position changes among bank employees. Some short term disruption and loan account manager turnover may therefore be expected.

Consumers value choice and SME business owners are no different in this respect. Thus, some SME owners change financial institutions on occasion. The decision to switch banks can be a costly one since a new relationship with a loan account manager must be established at a new bank branch. Nevertheless, SME owners do engage in `shopping', i.e. actively consider other financial institutions, and some do `switch', i.e. change financial institutions. Results reported by Riding et al (1994) indicate that both shopping and switching are influenced by the perception of the degree of competition between the banks. [30] The mergers, by reducing the number of banks, would reduce further the choices of many SMEs, especially those in rural areas, and lead to a reduction in their ability to engage in shopping and switching.

Table 5 below summarizes the main effects expected from these variables as they pertain to B.C.

Click here to view Table 5.

As can be seen from Table 5, the impact of bank mergers on SMEs in B.C. is likely to be negative. This conclusion is in accordance with the opinions of small business owners themselves. Results reported by the CFIB show that 64 per cent of their members oppose the bank mergers. [31]

What be the consequences for the B.C. economy of negative impacts on the SME sector of bank mergers? The answer to this depends upon the importance of the SME sector and the size of the predicted negative impact of bank mergers. While the latter cannot be predicted with any certainty, the importance of the SME sector can be readily demonstrated. Firms employing less than 50 people account for 98 per cent of the firms in B.C. and for 49 per cent of the number of firm employees. The smallest firms, i.e. those employing less than 5 people, account for 86 per cent of all firms in B.C. and for 20 per cent of firm employment. Furthermore, small businesses' rate of job creation exceeded that of large corporations by a ratio of 2:1 during 1983-93. [32] Given the importance of this sector to the B.C. economy any negative effects arising from bank mergers would obviously be unwelcome.

II.b.ii Low Income Borrowers

While small businesses in general face problems with access to capital this is exacerbated by low income entrepreneurs. For them, they may not have money to put into a business themselves or have sufficient collateral with which to secure a loan. The monitoring problems are therefore particularly great for these individuals and require a different approach to lending. In many cases, this has been met by institutions other than the banks. For example, the Calmeadow Foundation and Community Futures have been involved, with varying degrees of success, in providing micro-credit and loan guarantees to small business owners.

An example of a larger loan program is the one initiated under the auspices of the VanCity credit union in Vancouver which has recently committed $15 million over 5 years to the Self Reliance Program. [33] This program provides loans to unemployed clients of VanCity. The program, run in cooperation with West Coast Enterprise Division, uses a "character-based lending" approach. In this approach loans do not have a collateral requirement. Instead, individuals are screened by their own business plan, credit rating and two letters of reference. Loan managers also meet personally several times with the borrower to establish a personal relationship between banker and client. The costs of monitoring are typically higher and this reflected in the interest charged to borrowers. [34]

This market is not well served by the existing banks for the reasons identified above and has been taken over by other non-bank financial institutions. Mergers are therefore unlikely to have a major impact on lending to this group.

II.b.iii. Aboriginal Groups

Lending to aboriginal entrepreneurs has been affected by many of the same problems as lending to small businesses in general. However, there is an additional factor which comes into play here, namely, the Federal Indian Act which prohibits reserve land and buildings from being used as collateral. As a result, lending to aboriginal businesses has typically been carried out by specialist institutions using a variety of monitoring techniques.

Federally funded programs such as the Aboriginal Capital Corporations under Industry Canada are one such example. There are eight ACCs in B.C. Discussions with members of one of the ACCs indicated that lending in aboriginal communities relies heavily on the personal relationship established with the ACC loan manager (or equivalent), who is also usually an aboriginal, and on lending which is supported by the community leadership. In this way, communities themselves have a responsibility in overseeing loan repayment. Provincial government funding for aboriginal entrepreneurs is also available through the First Citizen's Fund and the B.C. Development Corporation. Non-profit organisations have also established lending programs. The Calmeadow Foundation, for example, has set up a First People's Fund which provides loans of between $300 and $3000 to aboriginal entrepreneurs using the borrowing circle peer monitoring mechanism.

This has been an area of lending in which banks have not historically been leading lenders. In the past decade, however, the banks have shown a much greater interest and awareness of aboriginal lending. Undoubtedly, a part of this is due to the Treaty Process which has the potential to change the financial resources and economic potential of many aboriginal communities. The banks have therefore become more active in establishing a lending profile for themselves in aboriginal communities often in conjunction with other institutions and agencies. Examples of bank initiatives in this area include: the appointment of Managers and Vice Presidents of Aboriginal Banking; the opening of bank branches on reserves; the CIBC has a joint lending program with the Business Development Bank of Canada for loans up to $500,000 for aboriginal enterprises; the partnership between the newly established First Nations Bank and the TD; the RB in the Northern Area of B.C. has started offering mortgages to employed aboriginals living on reserves with a band council guarantee being used in lieu of the more traditional collateral requirements. Aboriginal banks are also being established; the Peace Hills Trust has recently opened its first branch in B.C. in Kelowna.

This remains a banking services and loan market which is in its infancy as a commercial market although it is expected to grow significantly in the next decade. The banks have all begun to more actively target this market although the management of trust funds arising from Treaty settlements and the administration of band accounts is likely to be of more interest to banks than development lending to small businesses in aboriginal communities as such. These two factors - the low base and the growth potential - are likely to mean that the bank mergers, per se, are unlikely to have significant effects for aboriginal lending.

Section III. Market and Policy Responses

The analysis presented above points to the problems which may be caused by the proposed bank mergers in terms of access to financial services and access to capital. The analysis has been primarily static in nature, analysing the current situation and comparing this with a post-merger scenario. The trends found in the financial industry must, however, be seen in the wider context of economic restructuring that is occurring worldwide. If this context suggests that there are either counteracting or reinforcing pressures then these will need to be taken into account in formulating policy responses. For example, if there are strong counteracting forces then concerns over "financial exclusion" arising from bank mergers may be eased and the desired policy response accordingly revised. The term most frequently used to describe the wider context is `globalisation'. This is a nebulous term but nevertheless requires us to briefly consider its dynamic - it is indeed this dynamic to which the merger banks themselves refer as the prime reason why mergers are desirable.

Globalisation, as a process, has redrawn geography in many ways. One impact has been the rise of core cities and zones within the global economy and the marginalisation of other areas. [35] This has had implications for the autonomy of nation states as well as the spatial tensions within them. In response to the financial crises of the 1980s and 1990s Leyshon and Thrift (1995: p.312) argue that "new patterns of credit-creation emerge as money and credit is redirected away from poorer to richer (and therefore `safer') groups" and that "new patterns of financial infrastructure develop as financial institutions restructure their operations over space to bring them in line with these new flows of credit and debt." A similar restructuring has occurred as a result of globalisation as the spatial coordinates of the global economy increasingly become those of core zones and their wealthy inhabitants. Pressures for the spatial reorganisation of financial institutions are therefore present in the process globalisation.

Seen in this context, the tensions between the policy goal of full accessibility and the corporate logic of profitability and restructuring will mean that concerns over "financial exclusion" are likely to increase rather than abate over the coming years. It is this context that the analysis of section II of this paper should be seen. Similarly, policy responses must take into account the wider context as well as the specific findings of the B.C. case.

To summarise these specific findings, it was argued in section II that the proposed bank mergers do indeed pose challenges to the goal of full accessibility and, in some important respects, move us further away from it. In particular, it was been argued that:

Considering the results as a whole, the proposed mergers are expected to have a negative impact on access to financial services and access to capital in British Columbia.

In addition, responses are required to address the present situation. These responses may come from the banks themselves, from the market and/or from government in the form of a policy response. The banks themselves have sought to address some of the concerns identified above by:

Competitive market pressures have also led to other institutions playing important roles. The financial system in B.C. consists of a number of different organisations beyond the chartered banks. The banks play a relatively conservative role, focussing on retail banking and relatively low risk loans. The banking system does not play a leading role in venture capital lending or in lending to low-income groups. Other financial institutions, most notably credit unions, which are rooted more in local communities and by institutional design are more accountable to their members who are predominantly local residents have taken on some of the responsibilities of low income lending. This is often done in conjunction with non-profit agencies and the credit unions are also often actively involved in community development initiatives. Credit unions operate in communities where banks do not operate or have closed. In over 900 communities across Canada credit unions or caisses popularies are the only financial institutions. [37] In B.C. there are over 200 credit union branches and 32 communities in which the only financial institution is a credit union.

The trust companies also offer some competition to the banks (although some of them have been acquired by banks) and the trust companies association has pointed out that some regulatory changes may increase their ability to compete. In particular, they point out that regulatory reporting and compliance costs are higher for smaller institutions and have suggested that consideration be given to the establishment of a small institutions division of the OSFI. They have also noted that the proposed differential or risk-based premiums for deposit insurance may also have an adverse effect on smaller institutions. [38] Market pressures and institutional differentiation have therefore played an important part in providing a more accessible financial system. Consumer ratings of financial institutions rank non-chartered bank financial institutions more highly than the banks which indicates that if bank mergers do have a negative impact on consumers then non-bank institutions may increase their market share as a result. [39]

These competitive pressures could be enhanced by the relaxation of regulations which restrict foreign ownership of banks. The entry of foreign banks into Canada would, potentially, provide greater access to financial services and access to capital for Canadians. However, a number of important caveats are required here. Firstly, banking institutions strongly rely on reputation and it is only those banks with internationally established reputations which will be strong players in the Canadian market. These banks will therefore be the large multinational banks. This means that access to capital for small business is unlikely to be significantly improved since the monitoring costs of this sector for large international banks will likely be at least as high as those of the present large domestic banks. Secondly, increasing competition by foreign entry will primarily be beneficial to the higher income, urban consumers. Evidence for the U.S. reported by Amel and Liang (1998) indicates that new bank entry is much more likely (although even here infrequently) in urban areas than in rural areas. Their results indicate that new entry is primarily aimed at large and rapidly growing markets. This suggests that a major foreign bank presence can not be expected in small town and rural B.C. where accessibility to financial services is the most pressing concern. Indeed, as pointed out above, the global context within bank operations must be seen points to an opposite logic. Thirdly, branch banking requires significant physical infrastructure which suggests that new market entry by foreign banks could come about by a takeover of existing banks rather than as purely additional competitors.

The case for increasing competition in rural retail banking markets through easier foreign access is therefore one which is empirically weak.

The policy question remains, therefore, whether voluntary actions by the banks and market pressures from differentiated financial institutions (domestic and/or foreign) is sufficient to ensure access to a competitive financial system or whether other modes of regulation, emanating from government, are also required. The evidence suggests that at present accessibility is a major concern for the reasons identified in section II and that the voluntary and market responses, while important, nevertheless need to be complemented by government regulation.

Provincial and Federal governments in Canada have used a variety small business loan programs themselves as a way of increasing the capital available to small businesses. The federally funded Business Development of Canada and the Small Business Loans Act program being perhaps the best known examples. The B.C. government has a number of business start-up programs and participates in the Western Diversification sponsored loan programs. The Province of Ontario has also sought to influence bank lending behaviour to small businesses through the tax system and regulatory system. The 1996 Provincial Budget included a temporary surtax on the capital tax paid by the banks. This surtax could be earned back if the banks increased the availability of equity capital for small businesses. [40] The Budget also stated (1996: p.23) that "we will make sure that banks that do business with the Provincial government are also doing business with Ontario's entrepreneurs. Starting this year, banks that wish to provide services to the Province will be required as part of their tender bid to show a comprehensive record of small business lending." In addition, the Ontario government has introduced the Small Business Investment Tax Credit for Financial Institutions. However, the surtax provision will end in October of this year and, so far, only Canada Trust has launched an initiative under the latter program. Despite tax incentives, the major banks seem uninterested in participating in these programs since they view fiscal incentives of this kind as inappropriate.

The regulatory regimes established in a number of other countries are also of interest here. Perhaps the most well known is the that provided by the 1977 Community Reinvestment Act (CRA) in the U.S. Under the provisions of the CRA, banks are subject to reporting and being assessed on their lending activities. The reporting requirements and assessment criteria vary by bank size. Large banks (defined as those with assets over $250 million) are, according the latest revisions of the CRA, subject to three performance tests. These tests cover lending, service and investment. These tests include performance indicators directly relevant to the analysis of access to financial services and access to capital. Included are: [41]

The Lending Test:

The Service Test:

The Investment Test:

This provides a legal framework within which the banks' role in meeting the goal of accessibility can be assessed. As with any legal framework, the effectiveness of the CRA depends upon the implementation of the framework, for example, how stringent the performance tests are and how strictly they are enforced. In the U.S., the existence of the CRA has not prevented substantial "financial exclusion", and the effectiveness of any CRA-type legislation depends, therefore, on how it is implemented. [42]

This notwithstanding, this type of regulatory approach has a number of advantages in seeking to address accessibility. Firstly, it requires the banks to provide detailed information on their operations through data disclosure procedures. Part of the problem with undertaking the type of analysis conducted in this paper is that critical data is not publicity available. By requiring data disclosure a more informed public debate and public policy is possible. [43] Secondly, the CRA applies to all of the banks. Given the pressures for corporate restructuring and the pressures of `globalisation' in remaking economic geographies, the ability of any one financial institution to pay greater attention to accessibility is limited since competitors may not follow suit. For example, under a CRA-type regulatory regime all banks will have to find ways to effectively monitor loans to small businesses. By requiring all banks to meet the requirements of a CRA, the banks can be led to move towards a socially desirable optimum which could not be reached by private markets alone. Thirdly, and perhaps controversially, the CRA requires the banks to play a legislated role in providing accessibility. A concern has been voiced by Anti-Poverty Organisations that initiatives such as those associated with the Four Corners Bank in Vancouver will simply lead to a ` financial ghetto' as the mainstream banking institutions withdraw from the area and are being replaced by specialist financial institutions for low-income groups. A CRA can provide a mechanism to ensure that banks do not withdraw in this way but remain accessible to all groups. Fourthly, form a Canadian policy makers perspective, the existence of the CRA in the U.S. has provided it with a degree of legitimacy and acceptability among bankers. Indeed, the Bank of Montreal acted upon some of the requirements on the CRA since its acquisition in 1993 of the Harris Bank of Chicago. From a policy-makers perspective, a CRA-type regulatory regime is capable of being harmonised among the Provinces, with the performance criteria providing a uniform basis on which Provinces could judge banks in their tenders for public contracts.

Other policy responses are found in other jurisdictions. In France, for example, individuals have the right to a bank account. If an individual is refused a bank account by any two banks, the individual can go to the Banque de France which will then assign a bank for the individual. This approach develops the concept of "financial citizenship" and guarantees that all individuals can exercise such citizenship. [44] "Financial exclusion" is therefore addressed by recognising that access to financial services should not be viewed as a product like any other, to be decided by private market individuals, but constitutes a right, a necessary part of citizenship.

In some policy areas, such as the access to capital by low income individuals and aboriginal groups, the proposed bank mergers are expected to have only a limited effect since these markets are currently of limited importance for the banks. Other lending institutions and monitoring mechanisms are typically in evidence in these sectors. However, it might nevertheless be thought appropriate, as part of an overall policy response to the mergers, to consider these accessibility issues here. Performance criteria for lending could play an important role here as could incentives (either positive or negative) for the banks to develop partnerships with other agencies currently addressing these issues.

A summary of the barriers to accessibility and the possible policy responses is given in Table 6 below.

Click here to view Table 6.


View Report References


View Appendices Table 1 - Population, Average Income and the Distribution of Financial Institutions in BC Communities

View Appendices Table 2 - Population, Average Income and the Distribution of Financial Institutions in BC Communities with at least one merger bank (number of financial institutions)

View Appendices Table 3 - Branch Closure Vulnerability Index (BCVI) for BC Communities with at least one merger bank

Click here to view graph

Source: CFIB (1997: p. 10)


[1] I am grateful to Ms. Dana Preda for her excellent research assistance on this project.
[2] See The Chancellor Partners, (1997: p.8)
[3] As part of the B.C. Task Force inquiry a companion paper has been commissioned on the impact of bank mergers on consumer services. Readers are referred to this for more detail of the possible effects of increased concentration on industry conduct.
[4] The data is taken from Bank Of Montreal, 1998.
[5] See Vesala (1995). The Wallis Committee (1997: p.172) notes that in the Australian context "views in submissions were split as to the extent to which barriers to entry and exit remain and the extent to which those remaining constitute an impediment to the contestability of the market."
[6] See also Rhoades (1996) on this point.
[7] See Rhoades, (1997), p. 2
[8] The figures were 4153 and 1806 respectively. See Rhoades (1998: p. 3)
[9] The threshold is 1800. One third of MSAa exceeded this value. See Rhoades ibid.
[10] In a small number of cases, the community boundaries used by the financial institutions differs from that used by Statistics Canada. For example, the Canadian Payments Association data include Oak Bay as part of Victoria whereas the Statistics Canada population census data treat them as separate communities. Communities for which accurate financial institution data is difficult to compute because of differences in boundary definitions have been omitted from Table 1.
[11] Ideally, it would be useful to be able to weight financial institutions by their market share. Statistics Canada has assisted the Competition Bureau in developing a spatial analysis tool to examine multi-product mergers in a local market context. However, this was unavailable to the author. The Canadian Bankers Association has data on branch level sales. However, this too was unavailable to the author.
[12] As Informetrica (1998: p.5) notes "unquestionably there will be some branch closures and considerable consolidation of services".
[13] For a study of European bank mergers which casts doubt upon their efficiency enhancing effects see Altunbus, Molyneux and Thornton (1997).
[14] I am grateful to Dr. Tomson Ogwang for discussions on the problems of constructing the BCVI.
[15] See, for example, the debate over the construction of the Human Development Index reviewed in Noorbakhsh (1998).
[16] Research on branch closures in Britain indicates that there has been "a faster rate of branch closure in rural areas than in urban areas, since it is more expensive per capita to serve customers in rural than urban areas." (Leyshon and Thrift, 1995, p.317)
[17] Ideally, the population should be average figure over some time period. In order to match income and population data, a one year figure, that for 1995 is used for population. This has the disadvantage that changes in population from year to year may lead to changes in the index. The index should therefore be seen, strictly, as the index for one year.
[18] Leyshon and Thrift (1995: p.317) argue that in urban areas in the U.K. "the geography of branch restructuring reflects an existing pattern of income and class." This is supported by New Economics Foundation (1997).
[19] See CCRC (1997a).
[20] I am grateful to Dr. Fiona MacPhail for sharing her results on these trends.
[21] See CCRC (1997a).
[22] I am grateful to Kathleen Jameson of the Social Planning and Research Council of B.C. for the details of the financial situation of the Downtown Eastside.
[23] Electronic services includes ABMs and telephone and computer access to banks. The profit margins by electronic sub-group were not presented in the original report.
[24] CFIB (1997: p.7)
[25] The CCRC (1997c: p.7) state that "the small business sector, while creating over 35% of Canada's gross domestic product, is receiving at most 7% of the total business credit extended by the banks."
[26] See M.P. Tony Ianno's Quarterly Report on Lending by Major Banks to Small and Medium-Sized Businesses, May 27, 1998.
[27] CFIB (1998a: p.13).
[28] The Wallis Commission (1997: p.179) notes that "the Australian Business Chamber argued in its submission that `a go-ahead for mergers would spell a return to a "credit squeeze" on SMEs'".
[29] See, for example, Prasch (1992); Meyer and Reaves (1997).
[30] In an earlier study, Haines et al (1989) report that 30 per cent of business that did not shop for a new bank did so because their perception was that there were no real differences among banks.
[31] See CFIB (1997: Annex C). 23 per cent respondents thought that the bank mergers should be permitted, 11 per cent were undecided and 2 per cent expressed no interest in the issue.
[32] Data on small businesses is taken from Province of British Columbia (1996).
[33] I am grateful to Val Lockyear if the West Coast Enterprise Division for information on this program.
[34] The usual interest rate on loans is prime plus four per cent.
[35] See Castells (1993).
[36] For example, in 1994 the CIBC appointed its own small business ombudsman and the TD set up a three-party panel to address small business problems. See Democracy Watch (1994: p. 10).
[37] See Credit Union Central of Canada (1997), p. 13.
[38] See The Trust Companies Association of Canada (1997).
[39] A National Quality Institute Survey in 1996 rated credit unions third, trust companies ninth and chartered banks sixteenth of 21 industries in terms of customer satisfaction. See CCRC (1997c: pp.8-9). Credit unions were also rated highly by small businesses in studies conducted by the CFIB.
[40] See Province of Ontario (1996: p. 22).
[41] This list is taken from The National Community Reinvestment Coalition (1996). This document provides details of the full list of performance indicators and date disclosures.
[42]For a generally positive assessment of the CRA see, for example, Marisco (1996). It is also worthy of note that, according to Affordable Housing Finance (1998: p. 48), "NationsBank and BankAmerica have committed to a record $350 billion in loans and investment to underserved communities over 10 years under a proposed plan to meet their [CRA] obligations once the two banks complete their planned merger.
[43] For some of the problems of the data which are publicly available see, for example, CCRC ( 1997b).
[44] "Lifeline accounts" in the U.S. and "minimum accounts" in Germany are intended to play the same role.

Return to British Columbia Task Force on Bank Mergers Report Table of Contents

Return to Task Force Home Page

Last updated: September 4, 1998