Financial crisis in Asia in the late 1990s has highlighted the importance of money in economic growth and development. The essential issue is the amount of money available to an economy. When everything else is the same, an economy with an ample supply of money would experience more rapid growth than would otherwise. The amount of money available to a national economy or the supply of money is determined by the operation of the financial market as well as the monetary policy. Most analysts concentrate on monetary policy as a factor in determining the money supply at the national level. This paper deals with factors other than monetary policy, which may affect money supply at the regional level of a domestic economy. There has been a tradition in regional economics that examines determinants of the regional supply of money. (7) Geographers have discussed the command and control structure of banking. (5) In addition, both economists and geographers have discussed issues related to banking market structure. (6, 9) This research is set on a conceptual footing that market structure, and control structure of the financial industry affects money supply at the regional level. Consequently, market structure and control structure of the financial industry act as a factor that affects economic growth and development. Such a conceptual base requires an investigation that involve several related issues. First, what is the market structure and command and control structure in a regional financial market? Second, how does the market structure and control structure of a regional financial industry affect regional money supply? Third, how does the market structure and control structure of a regional financial industry affect economic growth and development? As an initial step toward study in this direction, this paper deals specifically with the first issue, the market structure and the structure of the command and control of a regional financial market. The study area is Kentucky. The financial industry in this paper is defined as all depository institutions, including commercial banks, thrifts (savings banks and savings and loan associations) and credit unions. The reason that Kentucky is chosen as study area is that the state contains economically drastically different counties, such as Appalachian, rural non-Appalachian, and metropolitan counties. This provides some "typical" examples for comparison. In addition, the reason that the depository institutions are chosen as the focus of study is due to the ease in obtaining information on the locations of operating offices.

Mainstream economic thought in the late 20th century has seemed to arrive at a worldwide consensus concerning the relationship between economic growth and development on the one hand, and money on the other. Increasing the amount of money to promote growth or tightening money supply to put a brake on inflation has become part of regular macroeconomics policies adopted by governments in the West and other places alike. Money mobilizes real resources that are in idle and channels resources into further services by functioning as rewards to resource owners. Money acts a similar mechanism at the regional level of a domestic economy. Regional economic stagnation or distress reflects a state of aggregate supply and demand that works against a region. On the supply side, a region may lack the resources necessary for a meaningful growth such as needed natural resources, physical capital that is embodied in various economic sectors, and a labor supply of sufficient quantity and quality. Problems within Appalachia have long been attributed to traditional economic sectors such as agriculture, coal mining, traditional manufacturing, etc. that have been in a long term decline since the early 20th century, to a lack of sufficient human resources such as education and entrepreneurship, and as well as to a lack of easy access. (8) Traditional development policies have mainly focused on the improvement of regional supply conditions, as evidenced by highway construction, community and education developments, etc. in various Appalachian development programs. (2, 3)

On the demand side, lack of sufficient demand in a region may be due to an insufficient amount of money related to low income, lack of viable investment opportunities, etc. (2, 3) The strategy to promote tourism in Appalachia can be seen as an effort to market the region as a profitable investment location for new industries. Little effort has been given to the discussion of increasing the money supply of Appalachia as a possible strategy to enhance the aggregate demand. It may be illuminating that the 22 million people in Appalachia in 1997 was comparable to the population of Malaysia, which achieved startling success in growth and development prior to the 1997 financial crisis. In terms of physical size, Appalachia's 200,000-square-mile area is slightly larger than Malaysia. The success of Malaysia in promoting growth is partly attributed to the massive inflow of foreign capital. It is entirely conceivable that an independent country of the size of Appalachia could achieve a rate of growth seen in Malaysia via attracting capital from foreign lands. The reality is that the chronic decline of Appalachia demonstrates a declining money income, which contributes to a lack of demand.

It may well be that Appalachia's status as an integral part of the United States has hurt its chance for growth. This is so because monetary policies designed to stimulate regional aggregate demand would be ineffective within a national economy where a single national currency is adopted. Under a single currency, absence of spatial difference in exchange risk would result in a redistribution of money from policy-target regions to unintended regions via correspondent banking and other forms of investment. High unemployment rates and out-migration have prevailed in many areas of Appalachia since early this century, reflecting a region with untapped resources. It was no help that the Federal Reserve gave up the practice whereby different regional reserve banks set interest rates according to the regional financial conditions. In reality, regions in slow growth may well need some extra amount of money to mobilize their resources, and regions with an overheated economy may well need a brake on their inflation. A uniform interest rate across the country set by the central bank would provide money with the same credit terms and may not fit various regional conditions. The result is to weaken the role of monetary policy as a viable tool in promoting regional growth.

Since it is unlikely that there would be an Appalachia oriented monetary policy, the monetary policy as an option in promoting regional growth seems out of the question. However, there are other factors that also affect the amount of money potentially available to a region. One of these factors is the financial market structure. The theory of the regional financial market (1, 4) suggests that there exists a spectrum of regional segmentation of financial markets, ranging from perfect financial capital mobility to complete segmentation. Financial segmentation refers to the imperfect mobility of financial capital between regional and national markets which arises partly due to a less than competitive institutional structure of the local banking industry. In a competitive financial market, a shortage of funds will translate into higher regional interest rates than in the national market, and thus attract national capital into the region and augment the credit supply. However, in a less than competitive market fixed-price tends to prevail via mechanisms such as price-leader. Fixed-price will result in the allocation of funds via rationing in which the "implicit contract" is struck between the lender and borrower to minimize the risk. The economically stressed regions or a local market unfamiliar to investors at the national financial market may be required to pay a location-related premium on top of the regular interest charge to correct the investment risk associated with the region. The fixed-price and implicit contract in less than competitive markets may force down such location-related premiums, and thus discourage financial flow from national to regional markets. The result is inadequate supply of credit in a local area.

Corporate control structure of the financial industry may also affect the regional money supply. (10) Entries by outsiders may change the market structure of the financial industry in a region. In addition, the presence of outsiders in a local market may exert pressures on the local institutions to increase transparency in business dealings and become more competitive. In discussing Asian financial problems, analysts have demonstrated a perplexing contradiction concerning the merit of outside ownership in the financial industry. On the one hand, many blamed foreign ownership as a culprit in causing the sudden withdrawal of money from emerging economies that eventually led to the financial crisis. On the other hand, prescribing remedies to the problems in the financial sector of emerging economies, many analysts have hailed foreign ownership in the banking sector as a way to increase the competitiveness of a country's financial industry. Although these conflicting arguments appeared since the Asian financial crisis, they are simply variants of the same conflicting statements that have made their appearance in the U.S. banking industry. There have been constant battles in U.S. banking history between banks with various locational affiliations such as community banks, regional banks, and banks with national dominance. The essence of the battle is the claim to bank resources lying in a community. Before banking deregulation, there had been a long tradition that saw bank ownership in a local area by outside banks as a "bad" thing. The reason was that outsiders would necessarily take local deposits and invest somewhere else, which would drain resources out of a community and thus hurt its growth. The agitation against branch banking and interstate banking, and strong sentiment against large city banks were partly based on such an argument. The recent two decades have seen a reversal of the tide in which banking consolidation has swept the United States in the name of increasing local market competitiveness as well as achieving economies of scale and scope. However, old arguments die hard. Today, there are still concerns that non-local banks may not be willing to lend to local establishments, especially small businesses. The Community Reinvestment Act (1977 and revised in 1995) is partly intended to make outside banks accountable for providing credit to small businesses in the communities in which they operate. Clearly, outsider ownership could be an evil when it is seen as the source of resource drainage, or virtue when seen as a way to improve the competitiveness of a local financial market. Since the logic could go either way, the impacts of outside ownership largely become a matter of empirical evidence.

In this study, all Kentucky counties are grouped into three categories. The first is Appalachian, defined as rural counties that are designated as Appalachian by the Appalachian Regional Commission (ARC). Metropolitan areas that are designated as Appalachian by the ARC are not included in this group. In 1997, there were 44 counties in this category. The second group are rural non-Appalachian counties. There were 54 such counties in 1997. The third group consists of 22 metropolitan counties, including 17 non-Appalachian metropolitan counties, and 5 metropolitan counties designated as Appalachian by ARC. In 1997, per capita incomes in all but two rural Appalachian Kentucky counties were less than 70% of the U.S. average. In comparison, only 2/5 of the non-Appalachian rural counties, and 4 out of 22 metropolitan counties, had a per capita income less than 70% of the U.S. average. In 1996, the percent of people in poverty in Kentucky as a whole was 17.9%. All rural Appalachian counties except one had a poverty population higher than the state average, more than half of which were higher than the state average plus one standard deviation. In comparison, only a third of the rural non-Appalachian counties, and 2 out of 22 metropolitan counties had a poverty population ratio higher than the state average. Outside rural Appalachia, only one county had a poverty population ratio higher than the state average plus one standard deviation. In 1996, the unemployment rate in Kentucky Appalachia stood at 8.1%, while the corresponding number for the U.S. was 5.4%, and for the Appalachia region as a whole 5.7%. In the same year, 15 rural non-metropolitan Kentucky Appalachian counties had a 2-digit unemployment rate. All but 7 rural Kentucky Appalachian counties are designated distressed counties by the ARC. In these distressed counties, per capita income is no more than two-thirds of the national average and poverty and unemployment rates are at least 150 percent of the national rates. Things are not changing for the better for the region while the American economic boom continues. In 1997, the mean unemployment rate for Appalachian counties was 8.7%, in comparison with 6.1% and 4.7% for rural non-Appalachian and metropolitan counties respectively. Apparently, during the current national economic growth cycle, the longest during the post war era, when many other regions in the nation face an increasing demand for labor, Kentucky Appalachia as a whole still suffers from a slack demand.

As of June 30, 1997, there were 459 depository institutions operating in Kentucky. These firms operated 1667 offices and owned $48 billion in deposits. A breakdown of these firms along the line of the type of institutions and their distributions in the three regions defined above are presented in Table 1. For the state as a whole, commercial banks are a dominant group of depository institutions. The degree of dominance by commercial banks appears more prominent when one examines the percent of number of institutions, percent of number of offices, and the percent of deposits in sequence. In comparison, thrifts and credit unions hold a significantly smaller share of the market than commercial banks, and their share declines in a sequence with the percent of number of institutions, percent of number of offices, and the percent of deposits.

The patterns illustrated above generally hold for the three groups of counties, but with important variations. Of the three groups of counties, Appalachia demonstrates the highest degree of dominance by commercial banks measured in all three percentages. In comparison, credit unions had an extremely weak presence in Appalachia, especially as measured by the percent of the amount of deposits. Metropolitan counties, on the other hand, has the lowest representation by commercial banks of the three regions, measured in all three percentages. In fact, their percentage in the number of commercial banks is actually lower than that of credit unions. Rural non-Appalachian counties as a whole attracted the largest number of institutions, mainly commercial banks. Its share for different types of depository institutions is similar to the state average, especially in terms of the percent of deposits. A location quotient is constructed for the three groups of counties via dividing the percent of deposits of a region by the state average, Appalachia had a value of 1.05 for commercial banks, but a value less than 0.1 for credit unions. Metropolitan counties had a location quotient 1.2 for credit unions, but 0.97 for commercial banks. Both groups had a location quotient 1.1 for thrifts. Rural non-Appalachia had a location quotient close to 1 for all types of institutions.

* including savings banks and savings and loan associations
** each credit union is assumed to be operating in one office
*** includes deposits and shares in credit unions
**** less than 1%
Sources: National Information Center, Federal Reserve Board; Credit Union Data, National Credit Union Administration; Summary of Deposits, FDIC. All data are for 1997.

In 1997, Appalachia, rural non-Appalachia, and metropolitan counties had 22.7%, 29%, and 48.3% of the Kentucky population, respectively. Using the population share as a benchmark, the banking resource distribution among the three groups of counties demonstrate uneven patterns. Appalachia generally had a smaller share of banking resources than their population share would indicate. However, the unevenness in the number of offices is lower than that in the amount of deposits, as illustrated in Figure 1. This pattern exists for all three types of institutions. This indicates that disadvantages in the Appalachian financial market lie more with insufficient funds than with inconvenient banking locations.

Metropolitan counties contained a share of deposits larger than their population share would indicate. However, the share of the number of offices for metropolitan counties was similar to their population share, though considerable variations exist among different types of offices. For thrifts, there was a share higher than population share, and for credit unions, there was a considerably higher share than the population share. These higher shares, however, were mediated by a share lower than population share in commercial bank offices. Apparently, the advantage of the financial market in these counties are mainly reflected in ample funds available rather than in convenient banking locations.

Rural non-Appalachian counties had a share of deposits similar to their population share, especially in commercial banks and credit unions. However, they had a share of offices larger than their population share, especially for commercial banks. This suggests that rural non-Appalachian counties as a whole do not face particular disadvantages in terms of the amount of funds available. In addition, they have an advantage in terms of adequate locations.

For credit unions, the extreme contrast between Appalachian and metropolitan counties is seen once again. Credit unions had an extremely strong presence in metropolitan counties, but were conspicuously lacking in Appalachian counties. This pattern can largely be explained by the close association of credit unions with professional occupations and large organizations, which would presumably be under-represented in Appalachia.

At the county level, the pattern revealed above basically holds. For example, only 3 Appalachian counties had a deposit share larger than their population share, and 12 had a share in the number of depository institutional offices larger than their population shares. The number of rural non-Appalachian counties with a share larger than their population share in the above two categories were 18 and 39, respectively. This paper does not investigate the pattern for the metropolitan counties since most metropolitan counties in Kentucky form a metropolitan market with out-of-state counties, which makes the comparison less meaningful.

An important measure of market structure is Hirfendal-Hirschman Index or HHI, which is the sum of squared percent market share of all depository institutions in an area. The higher the value of HHI, the more concentrated the market is. The largest value possible is 10,000, which occurs when there is only one firm in the market. In 1997, the mean HHI was 5352 for all Appalachian counties, 3884 for all rural non-Appalachian counties, and 2786 for all metropolitan counties. Figure 1 reveals a distinct concentration of counties with a higher HHI in the Appalachian area, and a concentration of counties with a lower HHI in metropolitan areas. It should be pointed out that there are rural non-Appalachian counties that are highly concentrated in terms of banking business. This, however, cannot conceal the fact that for the region as a whole, Appalachia is more concentrated than the other two regions. For example, there were 25 Appalachian counties with a HHI higher than the state mean HHI, while there were only 17 in rural non-Appalachian counties, and 2 in metropolitan counties. Of the 14 counties with a HHI higher than 7000, a benchmark used by the Justice Department as high degree of market concentration, 10 were in Appalachia.

The 459 depository institutions operating in Kentucky in 1997 belonged to 385 independent financial corporations. More than half of them, or 53%, were independently owned institutions (banks, thrifts, and credit unions), but controlled only 16% of deposits and 17% of offices. The rest 47% were bank holding companies, which controlled 84% of deposits, and 83% of offices. There were 22 out-of-state institutions controlling 40% the deposits and 34% of offices. Table 2 reveals how depository institutions headquartered in one region distribute their business across the three regions in Kentucky. As can be seen, depository institutions in each of the three groups of Kentucky counties did most of their business within their own group, reflecting the high percent of ownership in deposits and offices by a region's institutions in its own region rather than in the other regions. However, compared with institutions in Appalachia and rural non-Appalachia, metropolitan institutions did more business outside metropolitan areas, especially in rural non-Appalachian counties. Out-of-state institutions also seemed to favor metropolitan areas as their business locations. Their percent ownership in deposits and offices is significantly higher in metropolitan counties than that in Appalachian and rural non-Appalachian counties. Interestingly enough, they allocated proportions of their business in the three types of Kentucky counties similar to those by Kentucky metropolitan institutions.

Figure 3 illustrates the controlling locations within the three types of Kentucky counties. Appalachian counties was least influenced by out-of-state institutions of all three types of counties. In addition, Appalachia as a whole was less influenced by metropolitan institutions than rural non-Appalachian counties. At the county level, a similar pattern can be seen. For example, the average percent of deposits controlled by out-of-state institutions was 11% for Appalachian counties, but 13% and 37% for rural non- Appalachian and metropolitan counties respectively. In addition, the median percent of deposits controlled by institutions outside a region --including institutions both outside a region and outside Kentucky-- was 17% for Appalachian counties, and 17% and 48% for rural non-Appalachian and metropolitan counties respectively.

Controlled in        Controlled by	State of Kentucky	Kentucky  Appalachia	Kentucky non-Appalachia	Kentucky Metropolitan Areas
Kentucky Appalachia	6043/237* (100/100)	5565/215 (92/91)	127/6 (2/3)	351/16 (6/7)
Kentucky non-Appalachian	10481/394 (100/100)	376/15 (4/4)	9721/361 (93/92)	384/18 (4/5)
Kentucky Metropolitan Areas	12160/465 (100/100)	923/37 (6/8)	1525/62 (13/13)	9692/366 (80/79)
Out-of-state	19384/553 (100/100)	1114/56 (6/10)	2402/107 (12/19)	15868/350 (82/71)

* amount of deposits in $million/number of offices

Sources: see Table 1

This research finds that the depository institutions in Kentucky demonstrate distinct regional patterns in market structure and in the structure of corporate control. In terms of market structure, Appalachia counties in Kentucky are the most concentrated, followed by rural non-Appalachian and metropolitan counties. The disadvantage in Appalachian banking lies more in insufficient funds than in banking locations. In contrast, the advantages in metropolitan counties are in ample funds rather than in convenient locations.

In addition, rural non-Appalachian enjoy ample amount of funds as well as convenient locations. In terms of the structure of corporate control, Appalachian counties as a whole are least affected by outside institutions including out-of-state institutions as well as metropolitan institutions. Metropolitan markets are most influenced by out-of-state institutions. Rural non-Appalachian counties are in an intermediate position. All these observations suggest a more concentrated and closed financial market in Appalachia than in the other two regions.

The above conclusions point to the direction for the next phase of investigation that should focus on whether the growth problems in Appalachia relate to their more concentrated and closed financial markets, as suggested by standard macroeconomic theory. Specifically, the regional lending pattern should be investigated in the context of market structure and the structure of ownership. In addition, the regional growth pattern should be investigated in the context of regional lending pattern as well as the regional market and ownership structure. It may seem too early to offer policy implications at this stage of the investigation. However, given the fact that most Asian economies have finally started to turn around after significant financial restructuring, among other changes, policies that are designed to strengthen the financial market in Appalachia and make it more competitive and open may be a positive step toward solving regional growth problems.

1. Amos, Jr. O.M. 1992. The Regional Distribution of Bank Closings in the United States from 1982 to 1988. Southern Economic Journal 58 (3): 805-815.

2. Appalachian Regional Commission. 1997. 1997 Annual Report. Washington, D.C.: Appalachian Regional Commission.

3. Bradshaw, M. 1992. The Appalachian Regional Commission: Twenty -Five Years of Government Policy. Lexington KY: University of Kentucky Press.

4. Harrigan, F.J. and P.G. McGregor. 1987. Interregional Arbitrage and the Supply of Loanable Funds: A Model of Intermediate Financial Capital Mobility. Journal of Regional Science 27(3):357-367.

5. Lord, J.D. 1992. Geographic deregulation of the U.S. Banking Industry and Spatial Transfers of Corporate Control. Urban Geography, 13(1):25-48.

6. -----. 1990. Impact of Banking Acquisitions on Local market Concentration in Florida. Southeastern Geographer, 30 (1): 1-16.

7. Miller, R.J. 1978. The Regional Impact of Monetary Policy in the United States. Lexington MA: Lexington Books.

8. Raitz, K.B. and R. Ulack, 1984. Appalachia: A Regional Geography. Boulder CO: Westview Press.

9. Rhoades, S.A. 1996. Bank Mergers and Industrywide Structure, 1980-1994. Staff Study 169. Washington D.C.: Board of Governors of the Federal Reserve System.

10. Whalen, G. 1995. Out-of-State Holding Company Affiliation and Small Business Lending. Economic & Policy Analysis Working Paper 95-4. Washington D.C.: Comptroller of the Currency.