12.1 THE GROWING CONCERN WITH REGIONAL DEVELOPMENT

In the previous chapter, we gained insight into processes of regional economic development involving the initiation and transmission of changes. We saw, also, that economic change within a region is determined partly by external forces beyond the influence of parties within the region itself and partly by decisions and actions that can be taken by such parties.

The present chapter will take up the question of what directions of change are desirable or desired: the objectives of regional economic policy. We shall look also into the pathology of regional development: what situations arise in which there is an urgent need for corrective action. Finally, we shall look into the prophylaxis and therapy aspects of regional development policy: what appropriate means exist for influencing development in desired directions and how they can be used most efficiently.¹

These questions are definitely in the spotlight today. Vast amounts of talk and action, and a substantial amount of thought, are directed at urgent problems of regional development (including but not limited to the special problems of urban life covered in Chapter 13).

Such concern is relatively recent. As late as 1948 it seemed fair to state that:

Although governments have a large stake in the results of locational development, great power to influence that development, and a correspondingly heavy responsibility for influencing it in a socially desirable direction, few governments have ever followed any coherent policy in regard to location.²

The "few governments" referred to certainly did not include the United States.

But a radical change in thinking was already brewing. In Britain even before World War II, it had become clear that the depressed economic position of the northern and Welsh industrial areas presented an intractable problem, and controversy was rife on which national policies might or might not work. Since the 1950s, we have been observing with some frustration that the so-called developing countries do not seem to catch up automatically with the more advanced ones, even with continued and massive international assistance of various types.

Moreover (as was noted in the previous chapter) economic statisticians and historians who had been investigating interregional disparities of income within the United States found reason to question the inevitability of convergence. One basis of concern and of desire for better understanding and policies has been a realization that regional stagnation that depression can be quite persistent.

Throughout much of the 1960s, attention was drawn to pockets of poverty in what otherwise was thought of as the "affluent society." Today, we are less apt to view regions with low levels of economic growth or even those experiencing absolute decline in economic activity as being anomalous. Structural changes in the U.S. economy, which are reflected by the population shifts described in Chapter 11, have given traditional concerns new weight. These shifts have meant expansion in the "sunbelt" states and relative or absolute decline in many "frostbelt" states. They ‘have also stimulated discussion of programs associated with "reindustrialization," a term that has come to mean recognizing that the economic base of vast regions of the United States is jeopardized by stiff competition and technological change. Indeed, some pundits now speak of the "rust-belt" as including much, if not all, of the nation’s old industrial heartland.

Important problems must be confronted in the face of such change. Transition can be difficult, whether it is accompanied by the expansion or the decline of local economic activity. Further, it is rarely easy to identify the cause or causes of change, and therefore coping with its consequences and planning for corrective action are also made more difficult.

There is a distinctly urban dimension to many of these regional problems. The process of urbanization accelerated earlier in this century because of the declining relative importance of agriculture. Unemployment in urban areas is more visible and more unsettling for both the individual and the community than is rural underemployment. Further, the rapid shift of black population from rural areas to urban slums³ intensified this change; and along with a complex of other problems of urban adjustment, it vastly increased the number of urban areas calling for external economic aid. Problems of traffic congestion and environmental pollution (particularly in and around urban areas) stimulated a search for more rational use of space and resources.

These problems developed during the 1950s and 1960s, when virtually every major metropolitan area was growing. Now that a number of larger metropolitan areas are experiencing population decline, additional problems have become apparent. At the same time, rapid metropolitan area growth in the South and West has meant new pockets of poverty and urban distress in these regions.

Fiscal pressures on local and state governments in the United States are part of the picture too. For expanding areas, with increased demands for all kinds of public services, the principal revenue sources of those jurisdictions (primarily the real property tax at the local level) often do not keep up with rapidly rising demands. States and local communities are rightly fearful that higher taxes will drive away or deter business investment.

The same fear persists in regions characterized by decline; the demand for services does not fall proportionately with population. Often, the least mobile persons—those left behind—are most in need of public services. This as well as the substantial resources required to maintain the existing infrastructure of roads, bridges, and sewer systems put upward pressure on tax rates that threatens to place areas hard hit by structural change at further disadvantage.

As a result of these forces, there has been increasing though sometimes reluctant reliance on the more ample and flexible taxing powers of the federal government to finance local programs (such as education, health, and highways), and still more broadly to provide unrestricted grants to the states for use at their discretion. This channeling of public money through the national treasury naturally brings to the fore rival regional claims on federally collected funds, and the competition may be intense as national policy makers are themselves forced to reconcile diverse pressures for increased expenditures with slow growth in revenues. Thus the problem of just and efficient allocation becomes one of the utmost concern.

Still another factor arousing interest in the policy problems of regional development is disillusionment with the effects and objectives of the more naive forms of local and regional self-promotion. As more localities participate in this competitive game, more of the total effort is recognized as simply canceling out (that is, each community is driven to promotional efforts in self-defense by the activity of rival areas). And more and more questions are raised about whether growth itself is a sensible standard of community interest and objective of public action at the local level.

Next, it appears that there has been a significant shift in the attitude of the general public, and of most economists, toward population growth on a local, regional, national, or world basis. In the 1920s and 1930s, the American credo of the beneficence of population growth was unquestioned, and leading economists and statesmen were pointing with alarm to the perils of economic stagnation that would beset us if we did not get busy breeding more young consumers. Malthus’s gloomy nineteenth-century warnings were dismissed as a discredited fantasy.⁴

This attitude has changed considerably. In part, the change came from the frustration of seeing hard-won output gains in so many of the underdeveloped countries canceled out by mushrooming population growth. Meanwhile at home the postwar baby boom, the all too evident pressures of population growth in urban and outdoor-recreation areas, the generally inflationary bent of the economy, and the relatively high fertility of people low on the economic and education ladder all helped to undermine the venerable New World tradition of the blessings of increased population. Today, thinking and policy are much more directed toward welfare objectives, such as fuller employment and higher per capita income, rather than to the misleading standard of aggregate growth.

Still another contributing factor in the shift toward more enlightened approaches to regional promotion is what might be called the dilution of provincialism. We now find it normal for individuals to make their home in several different communities and regions during their lifetime,⁵  and for them to travel often and widely. This more varied exposure is conducive to more objective feelings about programs that may benefit one region at the expense of another.

Finally, there have occurred (and are occurring) a number of important changes in the factors determining location choices of producers and consumers. These changes, arising mainly from changes in technology and increased income and leisure, really underlie many of the developments already mentioned and have certainly played a significant part in the rethinking on regional development. These changes in location determinants have been mentioned in previous chapters and can be briefly recapitulated as follows:

1. In terms of linkages among industries and their sources of materials and markets, the cost of physical transport of heavy and bulky goods is less important, and increased importance attaches to the speedy and flexible transportation of high-value goods and above all to communication—that is, the transmission of intangible services and information.

2. Access to markets has increased in importance for most industries compared to access to sources of raw materials and energy sources. This trend reflects the increased variety and complexity of products, which increases, in turn, the importance of shopper comparisons, sales promotion, and servicing, and thus makes proximity to market more desirable. Increased complexity of products has meant also more stages of processing between the primary extraction of natural raw materials and the final consumer, and thus a higher proportion of processes not directly using natural raw materials.

3. More and more importance is attached to amenity factors such as good climate, housing, and community facilities, and access to recreational and cultural opportunities. This change reflects rising standards of income and leisure, the increased importance of white-collar employment, and the fact that industries in a dynamic growth stage require a high proportion of well-trained and educated people, who are in short supply and so can afford to be choosy about where they will live and work.

4. For some industries, there has been an increasing degree of dependence on various services locally supplied by other industries, institutions, and public bodies. Thus we hear more about the external economies of a location well supplied with such services and facilities. We hear more of the importance of an adequate regional or community infrastructure—supplying such things as local utility services, police and fire protection, schools, hospitals, reference libraries, and the like—as a necessary basis for development of profitable enterprises producing goods and services for outside markets.

5. For other industries, the advent of technological advances in electronics and computer equipment has meant that production processes which had involved numerous mechanical parts, and therefore the close proximity of potential suppliers, have been replaced by new processes dependent only on the availability of one or several microcircuits. The producers using this new technology, as well as the suppliers of electronic components, are each relatively more free to choose among alternative locations, as compared with their counterparts using or supplying older mechanical parts.

Concern, controversy, and experience have brought into focus some basic issues of regional development objectives and policy, to which we now turn.

12.2 OBJECTIVES

12.2.1 Individual and Social Welfare Criteria

The ultimate objectives of regional economic policy run in terms of promotion of individual welfare, opportunity, equity, and social harmony. It would seem obvious, then, that economic policy in regard to a region should promote higher per capita real incomes,⁶ full employment, wide choice of kinds of work and styles of life for the individual, security of income, and not too much inequality among incomes. The relative importance of these goals is, of course, something that economics cannot tell us. Each of us has his or her own values and can try through the political process to influence the objectives of social policy so as to reflect those values.⁷

The aspect of equity raises some difficult questions in connection with the application of these criteria to programs and policies affecting such diverse groups of people as the inhabitants of a region. Any action— such as spending public funds for improved services, subsidizing the establishment of new industries in the region, or imposing restrictive controls on land uses—is sure to help some people more than others and may well help some at the expense of others. There is general agreement, however, on the guiding principle of the so-called Pareto optimum,⁸  which says that a change is desirable so long as it helps somebody without hurting anybody else. In practice, some of the benefits conferred on individuals by a change (for example, building a new highway) can be taxed away from these beneficiaries so as to compensate those who otherwise would suffer by the change; and the real question is whether the Pareto criterion is satisfied after feasible compensatory transfers of this sort have been made.

This guiding principle is much easier to propound than to apply. The very essence of a region is interdependence of activities and interests, and these interactions become particularly crucial in a high-density urban region within a city or neighborhood. Any change in one activity produces externalities and neighborhood effects on a variety of other activities, and these effects can be either helpful or harmful. Thus the building of a sports stadium can help the merchants of an area by bringing in more visitors and purchasing power, while at the same time it can spoil the surrounding residential neighborhood by creating traffic congestion, noise, and litter.

An important task for regional economists is to devise ways of "internalizing" the externalities involved in regional change. Take, for example, a chemical plant whose operations pollute a river. The pollution imposes a variety of injurious externalities on other residents of the area. Thus other industrial plants and water-supply systems downstream will have to incur extra costs for water treatment preparatory to use; businesses based on recreational use of the river, or fishing, will suffer diminished patronage, higher costs, or both; and there will be a still broader injury to the community in terms of loss of recreational opportunity and amenity, and possible health hazards. In principle, it might be possible to set a fee or tax on the chemical plant to reflect all these social costs, whereupon the costs of pollution would become internal costs of the chemical firm. These costs having been properly internalized, or placed where they belong (that is, imposed on the party that causes them), the chemical firm will have to reconsider its profit calculus. It can (1) choose a different location altogether; or (2) invest some money in effluent treatment to reduce or eliminate the pollutant, and thus get relief from the special tax; or (3) continue the pollution and pay the tax, whereupon the community gets the money to use for downstream water treatment or for compensating in some fashion the various parties injured by the pollution. Any one of these three outcomes is, of course, preferable to the original situation in which the chemical plant’s activities imposed social costs borne by other parties. This holds true regardless of whether the polluting firm absorbs or passes on to its customers the added costs imposed on it.

We can also speak of internalization in the opposite case, in which some individual activity yields external benefits to other parties but cannot feasibly collect directly from them in return. In such a case, the socially optimum scale of this activity is greater than the scale on which it will be led to operate on the basis of its costs and returns. Internalization of the social benefits will then be in the general interest. This is the rationale for the granting of various forms of subsidies, inducements, and exemptions to activities that are believed to have beneficial external effects. Thus a chamber of commerce or a neighborhood merchants’ association may raise money from its members to help build a convention hall, park, or other facility that they believe will eventually help their business; or a municipality or a state may use general tax funds to subsidize new industries, or give them tax exemptions, on the theory that such subsidy is a sound investment for the taxpayers as a group.

12.2.2 Regional Economic Growth as a Goal

What has been said above applies to economic objectives and policies for the welfare of a group of people. But a region is not, except at an instant in time, a definite group of people—it is an area populated by a changing group of people. In any region of consequence, every day sees some new arrivals (by birth or migration) and some departures.

This continual turnover of a region’s population complicates the question of policy goals. What is to be maximized over, say, the next ten years? The welfare of the present inhabitants of the region, regardless of where they may be in ten years’ time? The welfare of those who will be living in the region ten years hence, regardless of where they are now? Should it be counted as a regional gain if some people move in whose incomes are above the regional average, so that the average rises with their advent? If so, should one of the aims of regional policy be the out-migration of its poorer inhabitants? Is a region improved if its population and total income increase at equal rates, with per capita income unchanged?

Our preferred objective for a region’s development depends, of course, on where we sit. In addition to differences of interest among groups within a region, there is an important difference between the optimum for any single region and the optimum pattern of regional growth rates in relation to national welfare.

Is simple regional growth (in aggregate terms, without regard to per capita income or welfare levels) a sensible objective? On the face of it, such a criterion sounds quite irrelevant. Yet in practice we find that regional promoters and governments spend a great deal of money and effort in the avowed pursuit of a bigger regional economy—that goal is put forward without apology as something worth striving for.

How can this be explained? Partly, perhaps, by emotion and tradition. The idea that "bigger is better" has been a remarkably enduring component of American ideology, although it is no longer such a universal article of faith.

There is an even more basic explanation, however. A substantial part of the business and political interests in a region are, in locational terms, oriented to the local demand and thus have a direct stake in the overall population and income of the region. Department stores, newspapers, banks, utility companies, real-estate owners and speculators, and local political leaders have vested interests in aggregate growth. Their fortunes depend not so much on how well off the region’s people are as on the size and growth rate of the population. Net in-migration is good from their standpoint even when accompanied by a reduction in per capita income and other aspects of individual welfare; population losses are viewed with alarm.⁹

Quite logically in terms of their own interests, therefore, these groups are active promoters of and contributors to any programs and policies that promise to expand the regional economy in terms of aggregate income and employment. It is they who most zealously support chambers of commerce and local or regional booster associations. Firms primarily involved in export business have little or nothing to gain from such participation and, indeed, often stand to lose (through higher costs of labor, land, and some other local inputs) in a community or region experiencing rapid growth.

Here we have still another important contradiction to the widespread view, discussed in the previous chapter, that the primary sources of regional growth lie in the exporting sector. Local promotional efforts, at any rate, come mainly from the local market-serving or nonbasic sector.

What has just been said refers to regional growth in aggregate terms. In terms of individual economic welfare in the area (as roughly gauged by per capita real income), the interest groups play quite different roles— and it is in such terms that civic responsibility should really be judged. The local market servers, in their pursuit of the gains to be had from population growth and added business and housing development, too often assume that what is good for themselves must be good for the community, and then proceed to sacrifice quality of life for quantity. Since these interests include such powerful voices as those of utilities, banks, merchants, local public officials and union leaders, and (most important) the local news media, they can easily push an area into destructive overdevelopment. In one area where residents had been protesting for years the frantic and planless replacement of orchards and pleasant countryside by solid square miles of subdivisions and shopping centers, the leading newspaper publisher defended his advocacy of still more growth by the frank observation, "Trees don't read newspapers.

Finally, we note here still another mechanism by which regional or community change can become self-reinforcing and cumulative. Rapid growth confers increased income, prestige, and political influence on real-estate brokers and promoters, builders, and the other groups whose interests are served by local expansion as distinct from improvement of local well-being. This added power helps "growth-at-any-price" pressure groups to shape local planning and policies toward still further emphasis on continued growth and still less consideration of environmental and other welfare effects. Pure boosterism is truly narcotic, producing first euphoria, then addiction, and eventually decay.

12.2.3 Regional Objectives in a National Setting

Regions are not self-contained nor independent of one another. Accordingly, a true concern for human welfare calls for evaluating development and framing policy goals on a multiregional or national basis.

National High-Employment Policy and Regional Economic Adjustment. Experience has taught us that we cannot expect any satisfactory solution to the problem of regional unemployment or arrested development except in the context of a prosperous national economy. In a depression period, businesses are doing relatively little capacity expansion and have little difficulty in finding locally the necessary labor, services, and space for such expansion as they want to undertake. Their investment is more likely to take the form of cost-cutting improvements in existing plants, and this may well involve closing down some branch facilities at the more marginal locations. Moreover, in slack times, the surplus manpower in any area has literally nowhere to go and fewer resources to go anywhere; we cannot look to labor migration for any significantly useful adjustment.

We have found also that the national monetary and fiscal authorities have great powers to increase the nation’s money supply and disposable income, and thus to stimulate spending and investment in the aggregate. Such action helps to maintain the necessary buoyant climate in which constructive regional adjustments by people and industries can occur.

Efficiency, Equity, and Structural Unemployment. Some people feel that maintaining a high level of employment and demand in the economy is as much as the national government should do in regard to regional economies. There are, however, two distinct arguments for other, and more specifically region-oriented, national policies and programs.

The first argument invokes the criterion of efficiency, claiming that there are other ways besides fiscal and monetary policy for facilitating effective allocation of resources among regions and the necessary dynamic adjustments. The second argument is based on equity,claiming that the national government has a responsibility for helping disadvantaged regions as such.

The efficiency argument rests largely on the idea of "structural unemployment." This type of unemployment comes about because there are wide disparities in the employability of different groups in the labor force. There are poor matching between the kinds of labor that are in demand and those that are available, and there is insufficient mobility and interchangeability within the labor force. This makes shortages, rising costs, and consequently inflation inevitable, while millions of the less employable are still out of work. Obviously, any policies that will reduce these wide disparities and make manpower more mobile and interchangeable will have the good effect of shifting the inflationary brink closer to the ideal of full employment.

There is, then, a strong case for public programs involving education and worker training and retraining, and for more direct aids to spatial and occupational mobility: for example, improved information about job opportunities, assistance to migrants, and removal of racial and other discrimination in employment. It is also clear that such efforts ought to focus on upgrading the least advantaged types of workers and reducing their competitive handicaps. Such emphasis is, of course, in accord with equity objectives as well.

Helping Regions and Helping People. When it comes to translating this policy into geographical terms, we pass from consensus into controversy. It is tempting to argue that if public policy should specifically help the less-advantaged classes of people to find jobs, then it should by the same token seek to underwrite the prosperity and growth of all communities.

Such a view has been aptly characterized as substituting "place prosperity" for the more fundamental objective of "people prosperity."¹⁰  its more naive expressions, the place prosperity doctrine represents merely false analogy: an unreasoning assumption that whatever is true of individuals must also apply to areas. On a more rational level, it is possible to suggest place prosperity as a pragmatic proxy for the ultimate ideal of people prosperity—on the hypothesis that the best way to help a person is to promote the overall prosperity of the area in which he or she happens to live.

The place prosperity doctrine will figure importantly in later discussion in this chapter. For now, it is enough to indicate two of its shortcomings. The first lies in ignoring the fact that a region does not correspond, for any length of time, to a fixed set of people. Since people have some mobility, the best way to help disadvantaged people who are living in a particular region may be to encourage them to move. Migration can, in fact, serve both the objective of efficient use of resources and the objective of interpersonal equity and distribution of opportunity.

A second criticism of the place prosperity approach is that in practice it is wastefully nonselective in its assistance. In any community or region where there are unemployed and needy people, there are also employed and prosperous people. Increased employment and income for the area as a whole may help those who need it most; but a large part of its local benefits will come to those who do not need it. Those surest to benefit, as suggested earlier, are generally property owners and the operators of established locally oriented business, such as utilities, banks, and commercial and consumer service firms.¹¹ Growth of aggregate area income and employment does not automatically mean improvement in per capita income or the reduction of unemployment, and it generally injures some while helping others. Such considerations suggest that attacking human hardship and lack of opportunity solely through place prosperity might be like using a shotgun to kill flies.

Regional Rivalry and the National Interest. The benefits of growth in a region are directly and strongly felt by certain influential interest groups, while the costs are likely to be more diffused and less well perceived. Most regions, consequently, devote some effort to furthering their own economic growth by attracting additional activities.

Regional rivalry, like other forms of competitive promotion and warfare, can be in large part self-defeating, or a "zero-sum game," contributing nothing to the national welfare. One region’s gain is another’s loss. This is especially likely when the regions are small and when the primary weapons are persuasion and subsidy. Resources such as capital and labor that are drawn to one area cannot be used in production elsewhere, and from a national perspective there is no net gain, unless the productivity of those resources is higher in the receiving region. From this perspective, the nation’s rate of growth is analogous to a pie; a bigger slice for one region means a smaller slice for some other region.

However, regional growth may be generative rat her than competitive. In this more positive light, efficiency gains in each region may promote national prosperity. As Harry Richardson puts it:

It is possible for national growth to be increased by faster regional growth, and it is possible for regional growth performance to be improved without additional resource inputs. Agglomeration economies and spatial clustering of activities may induce more output than if production is dispersed. Growth-inducing innovation may be adopted by local entrepreneurs, even though they were first introduced outside the region. A change in settlement pattern (i.e., a more efficient regional urban hierarchy) or a reorganization of the intra-regional transportation system may improve productive efficiency and promote faster growth.¹²

Thus enlightened local efforts to enhance a region’s growth potential can result in significant net benefits. These efforts may take the form of upgrading the region’s human and natural resources and public services, protecting and improving amenities, stimulating entrepreneurship and innovation, fostering cooperation among various business, social, and political elements, and discovering the true comparative advantages of the region for further development. All these effects favor better utilization of resources and are clearly in both the national and the regional interest. A logical national policy with regard to regional development should include some effort to channel the growth urge of regions into these constructive paths.

But it is also true that regional rivalry in development can be something worse than a zero-sum game if it distorts the efficient allocation of resources. This danger is inherent in the use of local subsidies, and most of all with respect to the use or abuse of natural resources and the neglect of externalities.

Competitive regional and urban development are clearly suboptimal. They may involve regions in a competitive race to offer up for private exploitation their air and water quality. The resulting resource deterioration involves transfer of income from local residents to business firms. Competitive tax concessions to attract development may also result in relative weakening of the public sector. Competitive regional development may involve serious external diseconomies resulting from failure to treat environmental units, such as river basins, as planning units. The larger the planning region, the more adequately externalities can be assessed.¹³

We see, then, that national policy in terms of the development of specific regions can help to achieve more efficient use of natural resources as well as to reduce regional unemployment and broaden human opportunity.

12.3 REGIONAL PATHOLOGY: THE EMERGENCE OF "PROBLEM AREAS"

Regions, like people, want a doctor only when they are sick. When a region is enjoying growth-euphoria and reasonably full employment, there is no great disposition to examine its situation and prospects in detail and search for ways to gild its robust health. National attention is directed only to those regions that are in trouble, and there always are enough of them to worry about. We assume, in other words, that in healthy regions the workings of the market economy under existing constraints are relatively satisfactory.

To focus on regional pathology is both politically and economically rational. Our diagnostic and therapeutic resources are limited enough, and we are more likely to find something helpful to do for regions with obvious ailments than to improve comparably an already good situation. The only risk is that we may thus overlook opportunities to nip unwelcome developments in the bud.

Our main concern here is with situations where things have definitely gone awry. Regional economic growth is not a smooth, straightforward process. The persistence of efforts to explain development in terms of successive "stages" attests to the existence of important discontinuities. We do not by any means know what all these are, how to foresee them, or how to deal with them. But we do know that the development of a region, like that of a nation, encounters from time to time crucial situations in which its future course can be significantly influenced by major planning decisions and policies. Alternative paths appear; one of the alternatives may be a further growth along some new line, and the other may be stagnation, arrested development, or even regression.

These crucial situations present the biggest challenge to our insight into growth-determining factors. The stakes are highest and the rewards for correct decisions, in terms of economic progress, are at a maximum.

12.3.1 Backward Regions

A familiar case is that of underdeveloped nations poised on the threshold of industrialization and threatened by a genuine Malthusian peril of overpopulation. Much effort has gone into defining the conditions necessary for a successful surmounting of the threshold, the so-called "takeoff into self-sustaining growth" process.

Most if not all of the advanced countries also include one or more backward regions, which seem to be hung up at a threshold on the road of development and not to have kept pace with the structural changes and the rising income and opportunity levels of the more fortunate regions of the country. In the United States, Appalachia, a huge zone characterized by rural poverty, straddles the Eastern Highlands from New York State to Mississippi. Other large areas demanding special developmental attention have been identified also, and many smaller pockets of relative poverty and apparently arrested development exist in still other parts of the country. In Canada, the extreme eastern part of the country (the Maritime Provinces) is regarded as the chief area of concern of this type; in Italy, it is roughly the southern half of the country (Mezzogiorno); in Sweden, it is the far north.

12.3.2 Developed Regions in Recession

A second and quite different type of problem area is the mature industrialized urban region afflicted by stagnation. In Britain, the industrial areas of southern Scotland and Wales and northern England entered this phase in the 1920s. In the United States, at about the same time, migration of the textile industry to the South laid heavy blight on the industrialized region of southern New England, and real rejuvenation with new industries did not set in for more than twenty years. In the Pittsburgh region, slow growth or decline in the leading industries caused fears of stagnation and regression that gave rise to a major community effort to reverse the trend after World War II.¹⁴

Symptoms of this particular syndrome are easily recognizable. The ailing region’s rate of growth has been increasingly subnormal for many decades. Unemployment is high and chronic. Out-migration is heavy. The area appears to have somehow lost the dynamic growth character that had brought it to its peak importance in days gone by. There is a feeling that unless something really decisive happens, stagnation will prevail indefinitely.

Such a situation can arise in a region whose economy is heavily based on a few activities that have themselves ceased to grow or have begun to decline. They are the activities of yesterday and today, but not those of tomorrow. But arrested growth in a region may also mean simply that the factors of interregional competition, in specific activities, have taken a trend adverse to that particular region. The region’s difficulties are compounded if both of the above conditions apply, so that it finds itself with shrinking shares of declining activities.¹⁵  An excellent example of this is the "Steel Valley," which encompasses much of the upper Ohio River Basin and includes such cities as Youngstown, Wheeling, and Pittsburgh. The westward movement of the market for steel has left this region with old technology and a declining share of total U.S. steel production; all this in an industry that finds itself at an overall disadvantage when competing with foreign manufacturers.

But in diagnosing the ills of such a region, it is not enough to determine the extent to which it is losing ground to other areas in major activities, or the extent to which its activities are no longer of the growth-industry type. After all, we could hardly expect that every activity would continue to grow forever, or that any given region could forever retain or increase its relative position in its principal activities. A healthy regional economy can absorb losses in its stride and shift its resources into new fields, getting a share of the emerging new rapid-growth activities to balance the inevitable decline of other activities.

It is important to keep this in mind when trying to determine the proper role of federal and local policies in regional development. Change is a necessary aspect of growth, and it is as inevitable that some regions will prosper and others will not as it is that some individuals will fare better than others. Nevertheless, when change affects broad areas of the country—as is presently the case in the United States—large numbers of people are involved, and the political pressure for a response to related problems may become intense.

However, all affected regions are not equally in need. For some, the basis for rejuvenation may have been established well before decline becomes evident, and the proper role of policy may be limited to easing the transition. Other regions fail to make such adjustment successfully, and we must ask why. Perhaps it is simply because the degree of specialization in nongrowing activities was so intense. Perhaps it is because the loss of competitive advantage in some important activities has been so drastic. Or perhaps it is because the region has developed a sort of economic arthritis that inhibits its ability to adjust to rapidly changing conditions.

Whether regional analysts operate as full-fledged physicians ministering to the economic ills of sick regions, or more narrowly as diagnosticians, they have a special concern for cases in which the patient seems deficient in resistance to infection and in ability to recover. We have to look beyond the immediate symptoms to the less obvious organic difficulties.

12.3.3 Excessive Growth and Concentration

In both types of problem regions thus far mentioned, a basic symptom is that employment opportunities have not developed (in amount, in variety, or in both) fast enough to keep pace with the size and aptitudes of the labor force. Resources are underutilized. Somewhat the opposite situation prevails in regions that undergo extremely rapid growth involving massive inward migration. The growing pains of such regions are felt as impairment of the quality of services, destruction of local resources and amenities through overuse, a high rate of obsolescence of facilities, neighborhoods, and institutions, and a general deterioration of the quality of life. The forestalling or mitigation of these effects through analytical foresight and advance planning poses a major challenge to regional specialists.

The most widespread and obvious present-day examples of the perils of too rapid development appear in two types of areas. One is the suburban fringe of metropolitan areas, where many factors have combined to produce sudden and often unforeseen growth. The other type of area comprises zones of special recreational amenity such as beaches. The growth of population plus its increased mobility, leisure, and taste for outdoor pleasures add up to a formidable threat to our basically nonexpansible resources of open space, clean water, and privacy. This problem obviously involves much more than temporary "growing pains." As was suggested in section 12.2.2, the pressures of interest groups in a community or region lend themselves to overemphasis on growth per se, all too often at the expense of well-being.

Related to, but distinct from, the question of too rapid growth is the problem of excessive spatial concentration of development, specifically in gigantic metropolitan centers. Concern on this score is felt in nearly every country. In the less developed countries, the problem is seen as exclusive concentration of modern industrial development, business, and population in the chief city. In France and England, the concentration of growth in Paris and London has been officially deplored, and attempts have been made to combat associated problems for a generation or more.

The question whether our large metropolitan areas are "too big" defies any easy answer.¹⁶  Part of the difficulty lies in the variety of possible criteria. Large cities have been variously assailed as hotbeds of vice, breeders of psychological and political disorder, and hazards to health and safety; and they have been extolled for equally diverse virtues. With respect to economic criteria, it is often argued that the rising costs of housing, public services, and similar items make large cities uneconomical as places to produce or to live. These diseconomies of size are said to outweigh, in very large cities, the positive advantages of urban agglomeration that we discussed earlier.

A strong substantial body of empirical evidence suggests that there are strong net economies in the provision of infrastructure and public services of middle-sized cities as compared to small ones. The curve relating per capita expenditures on items in these categories to city size flattens out somewhere in the 100,000-to-500, 000 population bracket, with a possibly rising trend thereafter.¹⁷ On this limited basis, there is no "economic optimum size" of city, though we might refer loosely to a "minimum efficient size."

There are difficulties with this approach, however, in that expenditures reflect differences in the quantity and quality of services provided as well as costs. Thus persons in large cities (where, as mentioned in Chapter 10, we expect to find higher real income per capita) may have demands for public services that are different from those of persons in smaller cities, and this will affect per capita expenditures. John L. Gardner has undertaken an analysis of municipal expenditures that accounts for variation in income and wealth across cities.¹⁸  For a wide range of expenditure categories, he finds that costs per family increase with city size. However, Gardner also finds that costs typically decline as population density increases.¹⁹ Thus it may be misleading to concentrate on size alone; the efficiency of cities appears to depend on population size and density jointly.

In any event, costs of public services are only one element in the comparative economic advantages of different sizes of cities. A more accurate approach to this problem would recognize that the activity of cities includes the production and consumption of private as well as publicly provided goods and services. In order to make valid comparisons, one must account for the incremental benefits and costs associated with each as city size increases.

Many regional economists see hidden disadvantages in very large cities, justifying a public policy of diverting growth from such cities to medium-sized ones. They argue that there are important external diseconomies (such as added costs of housing, congestion, and environmental spoilage) that do not enter into the calculations of the firms or individuals who contribute to city size by establishing themselves there—in other words, these costs should, but do not, work to limit urban growth. For example, an additional urban freeway commuter adds to congestion and causes losses to all the other commuters whom he slows up, but he does not have to pay for the added costs inflicted on the others and is not deterred from rising the freeway.

Such externalities are real enough, and we shall have occasion to consider them further in Chapter 13. But their existence does not necessarily imply a net bias toward excessive city size, as is frequently alleged. First, the usual argument assumes too readily that the external diseconomies of large city size outweigh the external economies; however, as we saw in Chapter 5, the economies associated with urbanization may be substantial. Further, it implicitly assumes that the adverse externalities fall on parties that have no recourse—that is, they are "locked in" and can neither leave the city nor raise the price of their services in order to compensate themselves for the injuries suffered.

By and large, this assumption is unwarranted. Individuals and firms subjected to such external diseconomies as air pollution, traffic delays, long commuting journeys, high taxes, expensive housing, or noise can (and do) decide that they will not stay in such an environment unless they are paid extra to do so. Urban populations are characteristically mobile, and pay rates do run higher in large metropolitan areas than elsewhere, as we saw in Chapter 10. This suggests that at least some of the disutilities that urban life imposes on the individual are being passed back to employers in the form of higher wage costs. The effect of the cost increases on prices is undoubtedly greater for local goods and services than for those traded between cities, since prices tend to be set in the national market for traded goods and services;²⁰  nevertheless location decisions will be affected.

The market forces set in motion by compensatory payments to labor for urban disamenities may not fully offset the tendency for cities to become "too big," but they certainly work to counteract that tendency.²¹  The extent of this offset will depend on the reaction of affected parties to externalities. If workers are immobile (or, more generally, if location-fixed resources are affected by externalities), they may not be compensated for urban disamenities. Similarly, if producers of traded goods lack mobility and also are limited in their ability to pass compensatory payments along to customers in the form of higher prices, the market adjustment to externalities will be incomplete. Thus the greater the mobility of workers and producers, the more we can expect that diseconomies will be internal to the city as a whole in that they fall on firms and households whose decisions affect city size.²²  This does not imply that we may dismiss concern about adverse externalities as such, or concern about the many serious problems attending urban growth, which do in fact tend to be most aggravated in very large cities.

The foregoing discussion suggests that the search for an "optimal" city size may be elusive. The task is made even more difficult by the failure of most researchers to account for the fact that each city is but one element in a central-place hierarchy. In Chapter 8, we found that cities specialize by function—ranging from the smallest hamlet to large wholesale-retail centers. This specialization was influenced by efficiency considerations: agglomeration economies encouraged some activities to locate in proximity to suppliers or potential customers. As a result, the mix of goods offered by trade centers varies, and large centers are characterized by a more complete set of activities. For persons residing in all but the largest urban area, some shopping in higher-order centers is dictated by the spatial organization of production. It follows that the most efficient size for a city depends on the array of goods and services provided elsewhere in the urban hierarchy and on the efficiency of transport and communications among cities.²³

12.3.4 Comparison of Characteristics of Problem Areas

Table 12-1 summarizes the results of a tabulation of American "problem areas" (mainly on a county-by-county basis) made by Benjamin Chinitz in 1967. The categories are along lines already suggested above, but for one striking difference. It is indicative of the dramatic change in our attitudes toward regional development that Chinitz’s category I (high income, fast growth) was as recently as 1967 considered a problem category on the basis of unemployment, with no mention of the environmental impact, destruction of amenities, and deterioration of the quality of living, which we now consider the major penalties of excessive growth. A substantial number of cases of fairly severe unemployment do occur from time to time in basically flourishing labor markets, such as San Diego in Chinitz’s sample. Often these are transitory situations reflecting cutbacks in federal defense-contract employment in the area, and in some cases the unemployment is mainly seasonal; but it may be more chronic in areas that attract large numbers of migrants by their amenities.

12.3.5 Regional Structure and Economic Health

Both a region’s growth and the quality of opportunity it offers depend not merely on external influences and location but also to a large extent on the mix of activities that the region has. Some of the relationships are simple and obvious, others less so.

As explained in some detail in Appendix 12-1, it is possible to separate statistically in any time interval the component of a region’s growth that reflects the activity-mix of the region from those components that reflect overall national growth rates and changes in the region’s competitive position. Other things being equal, a region will grow faster if it specializes in "growth industries," just as it will tend to have a low wage level if it specializes in low-wage activities, or a high skill level if it specializes in high-skill activities. But shift-share analysis does not really tell us much about why regions grow or improve. It says nothing about the important question of how a region’s ability to hold its share of existing activities or to attract new ones is affected by the region’s economic structure. Here we need to look into some less simple and obvious relationships.

Regional economic balance or, in somewhat more definite terms, diversification, has for a long time been viewed as a "healthy" structural feature worth striving for. The grounds for this view, however, have not been clearly articulated.

Thus it is sometimes assumed that a region with a diversified structure (many different kinds of activities and an absence of strong specialization) is necessarily less vulnerable to cyclical swings of general business conditions and demand. Actually, this is neither true nor logical, as was shown quite a long time ago by Glenn E. McLaughlin.²⁴  Diversification per se is roughly neutral in its effect on cyclical stability. What really makes a region especially vulnerable to cyclical swings is specialization in cyclically sensitive activities (mainly, durable goods industries and especially those making producers’ equipment and construction materials and components). Thus a specialized steel-making center such as Youngstown naturally has greater cyclical ups and downs of employment than either tobacco-processing centers such as Winston-Salem or Durham or a broadly diversified manufacturing center such as Philadelphia. Analogously, a community or region highly specialized in seasonal recreation (such as Virginia Beach or the coast of Maine) shows much more seasonal variation in employment than the average area, while a region specialized in some nonseasonal activity may be more seasonally stable than the average.

It is a different story, however, when we consider stability and other desirable attributes over a longer period. In time, any of a region’s activities will suffer arrested growth and perhaps decline or even extinction, either because the product itself becomes obsolete (as in the famous case of buggy whips, which sorely affected Westfield, Massachusetts, the principal whip-making center of the country) or because the region loses out competitively (as, for example, New England lost to the South in textile manufacturing, and Pittsburgh in the nineteenth century successively lost out as a leading producer of salt, wagons, cotton textiles, and refined petroleum products).

If a region is narrowly specialized, such a loss can be, at least temporarily, disastrous; in a diversified region, it is unlikely that a major proportion of the total activity will suffer at any one time. Equally significant is the fact that a narrowly specialized region is likely to show less resilience in recovering its stride by developing new activities to take the place of those lost.

This attribute of resilience is an extremely important aspect of regional economic health. It depends to a large extent on diversification, since diversity of employment develops a wide variety of skills and interests in the labor force and also among business entrepreneurs, bankers, and investors, and a wider array of supporting local business services and institutions. In such a setting, there is clearly a better chance for new kinds of business to get a start and to survive the hazardous years of infancy.

Diversity is not the only factor affecting resilience. The inhibiting effects of high specialization are compounded if the region is specialized in activities characterized by large producing units, large firms, and absentee ownership. Such large units are relatively self-sufficient with respect to most kinds of business services that smaller units tend to buy from others; consequently, a region heavily specialized in, say, steel making fails to develop a broad base of such supporting services. In addition, its business leaders and sources of local finance have a more restricted outlook and interest. The range of local external economies is underdeveloped, and the whole climate for new and small businesses and new lines of activity is much less favorable than it is likely to be in a region of similar size where the firms and production units are smaller, more numerous, and less self-contained.²⁵

Finally, a region’s resilience partly depends on the amount of overall growth momentum it has at the time the loss is experienced. If the rest of the region’s activities are growing vigorously, even a sizable loss may produce only a short spell of abnormal unemployment. Fluctuations from a sharply rising trend may not involve much absolute decline; distress is most meaningfully measured in terms of how long and how far the region’s employment is below the previous peak, rather than how long and how far it is below a trend line.

Moreover, a region that has been growing rapidly has a number of characteristics favoring resilience. The labor force is relatively young because much of it has been recruited through recent migration, and young adults move the most readily. Thus the labor force is likely to be more occupationally mobile and adaptable, and less afflicted by seniority and tradition. The same applies to employers. Facilities are newer. A greater proportion of the population has had the broadening experience of living in other places. There is a more buoyant community climate of expectation of growth and favorable change.

Such considerations as these help to explain why Pittsburgh, for example, took in its stride the losses of such important specialties as textiles, vehicle manufacturing, and oil refining during its dynamic growth period in the nineteenth century, but was very slow to recover from losses of preeminence in such specialties as steel, glass, electrical equipment, and coal mining after about 1920.²⁶

12.4 THE AVAILABLE TOOLS

Mention has been made of some of the ways in which a region can influence its structure and development from within. Also, it was suggested earlier in this chapter that a national government can do a great many things to assist healthy regional adjustment and development, even without having to make any decisions as to which regions should be favored or why. In general, help of this sort involves the provision of information and the improvement of the quality and mobility of productive resources—including labor, capital, and land. Aid to education and vocational training, improvement of communications and money markets, preparation and distribution of statistical and technical information, improved labor market information and placement services, and a wide variety of other programs help to reduce the structural underutilization of labor and other resources in all regions. We also noted that maintenance of a high national level of demand makes it easier for labor and capital to find their most productive uses.

Many national governments nowadays take the important additional step of designating certain regions for special attention. In a few special cases (for example, Greater London, Paris, and some recreational areas such as the National Seashores in the United States), the purpose is to restrict further private development in an area judged to be overcrowded. Much more often, the immediate purpose is to increase employment and income in a backward or otherwise "distressed" area. Let us have a quick look at some of the means that can be used for such ends.

One line of action involves easing the supply of capital to encourage growth of employment in an area. Federal, state, and local funds are made available at low interest rates, generally on a matching basis, to establish or expand business facilities. A wide variety of tax exemptions and incentives (such as deferment of taxes, allowance of larger write-offs against income before taxation, and special low assessments on real property taxes) further encourage private investors. Public authorities (often working through local development associations) also encourage business expansion in certain areas by direct investment involving the purchase and assembly of land, clearing of sites, and construction and operation of "industrial parks" provided with all the necessary utilities and sometimes with buildings that can be adapted or leased by private firms.

In the contrasting case of areas in which development is to be restrained, public policy is implemented by imposing restrictions on further private investment or land use.

Another policy lever involves transport costs and services and the construction or licensing of new routes. In regulatory decisions on freight rates, the regional effects are given some weight, and the regions that stand to gain or lose by the decision often mobilize impressive and costly efforts to protect their interests. Both private and public leaders in the Pittsburgh region, for example, battled persistently and effectively against Buffalo and Youngstown in favor of adjustments in freight rates on flour-mill products and against the construction of a canal connecting the Ohio River with Lake Erie. Authorization for United States participation in building the St. Lawrence Seaway was preceded by decades of controversy, with different regional interests aligned pro and con. More recently, many cities along inland waterways have been involved in efforts to attract federal assistance for the rebuilding and upgrading of locks and dams. They have also fought hard to promote the continuation of pricing policies that shift the maintenance costs of these facilities to the general public by avoiding the imposition of user charges.

Another tool is the regional allocation of procurement contracts (particularly the defense contracts of the federal government).²⁷  The procurement agencies themselves are not particularly interested in conferring regional stimuli except as a way of pleasing influential congressmen; but they have, from time to time, been adjured to follow policies of greater decentralization, or of preference to areas of high unemployment. A region especially can sometimes effectively increase the demand for some of its products by sales promotion in outside markets or protective measures designed to restrict imports, and some states have been quite ingenious in setting up interstate trade barriers for certain commodities, such as milk.

A region can sometimes be effectively aided in development by subsidized technological progress or technical assistance leading to more efficient and profitable ways of using some special regional resource. Thus federally supported research on new uses for coal may play a significant part in improving the economic status of Appalachia. In such types of research and development efforts, the state governments, universities, and private foundations in the region are generally active as well.

A region’s development can also be guided along more effective lines through support of general analysis of the region’s economic situation and potentialities and through the formulation of integrated development plans. Modest but significant amounts of federal funds and technical assistance have been made available for planning activity and demonstration projects.

Allocation of federal funds to improve local public services and utilities has been a substantial element in regional assistance, particularly in Appalachia. This includes, in addition to schools, health services, and roads, the construction of water supply and sewerage facilities, libraries, and some kinds of recreation facilities. The Tennessee Valley Authority operation, instituted in 1933, represents one of the earliest large efforts to use federal funds systematically to develop a particular region; the project emphasized control of water resources and electric power but also embraced a wide variety of other forms of development assistance.

Finally, and probably most important, are programs to upgrade and mobilize human resources through education, vocational training and retraining, easing of ethnic discrimination and other kinds of restrictions on employment, and assistance in job finding and relocation in search of employment opportunity. Such programs were mentioned earlier as being in the national interest in all areas; but the need for them is obviously greater in regions where skills and mobility are particularly restricted and where there is a particularly poor match between labor supply and the demand for labor.

12.5 BASIC ISSUES OF REGIONAL DEVELOPMENT STRATEGY

As soon as a national government assumes responsibility for the geographical impact of its actions, it needs to decide which areas merit its favorable attention. The answer is inevitably determined in part by political pressures, but it is clearly in the national interest to formulate and apply some more objective social and economic rationale.

We note an interesting shift in the use of terms to describe areas to which national public development assistance programs are directed. In the 1920s and 1930s, the British used to refer to their "depressed areas." Later, these same objects of solicitude were rechristened under the curiously neutral term of "special areas." Still more recently, they have come to be referred to as "development areas." In the United States in the 1930s, we used to refer to "problem areas," or "stranded areas"; later, to "redevelopment areas"; and now to "development areas," and to "growth centers within them. As to our less fortunate brethren across the seas, we used to refer to them as simply "poor" or "backward," or "low-income" countries. Later they became "undeveloped" and then "underdeveloped." Nowadays, it is considered more tactful to speak of the "less developed" or, better still, the "developing" countries.

Does this curious trend reflect anything besides euphemism—that is, a growing squeamishness about offending the sensitivities of people in the areas in question? Not necessarily; but perhaps we can read into the new terms a growing emphasis on the positive, and a belief that any and every region can and should be made to develop faster. We may descry also a disquieting indication that the ideal of place prosperity is enlisting greater support.

12.5.1 The Four Issues

Actually, three more strategy issues come to light here in addition to place prosperity versus people prosperity. One is whether we consider aid to regions as charity or as investment. Should we select areas on the basis of the greatest degree of distress (what has aptly been called the "worst-first" rule of priority) or on the basis of how much additional income and employment opportunity can be generated per dollar of aid? A further issue involves the spatial focusing of aid to areas: that is, what size area is a proper "development unit," what is the role of urban focal points within such areas, and should aid be concentrated at a few points or widely spread? The final issue concerns the appropriate choice of means of assistance from among the large variety of available devices sketchily catalogued in the previous section of this chapter.

These four issues (place prosperity versus people prosperity, distress versus development potential, concentration versus diffusion, and the choice of means of assistance) will recur often in the discussion that follows. We shall find that they are closely interrelated, and that none of them can be resolved as categorically as the word "versus" might imply.

12.5.2 Should Jobs Move to People, or People to Jobs?

If manpower is scarce in some areas while jobs of similar types are scarce in other areas, the situation can presumably be improved either by moving some jobs or moving some people or both. Both kinds of adjustment do take place spontaneously, though not by any means to the extent that would be necessary to eliminate or equalize regional structural unemployment. Both can be assisted or impeded to some extent by public policies. The question of which policy should be emphasized is a perennial one and was debated with particular heat several decades ago when the British government was trying to decide what to do about certain depressed industrial areas. It is a crucial question today in every country that is seeking to improve regional adjustment, and it particularly involves the two issues of (1) people versus place prosperity and (2) need versus development potential.

The answer depends on our judgments about the footlooseness of people on the one hand and that of investment and employment opportunity on the other. If we believe that people are reluctant to move, that we should not try to induce them to do so, and that practically any populated area can be made attractive to new employers, then it follows that the proper policy is to induce more employers to move to regions where unemployment is high. Consistent with this view is an emphasis on degree of distress as the criterion for allocation of assistance to regions, since it is assumed that people have to be helped in situ and that every region has adequate development potential. By this approach, place prosperity is equivalent to people prosperity. Finally, this view would imply that assistance should be given to individual small areas and should be widely diffused, since people are assumed to be tied to their labor market areas. To sum up, the elements of this position are: place prosperity, allocation on the basis of need to a large number of quite small areas, and inducements to employers as the principal means of assistance other than straight charity.

If on the other hand we judge that people can reasonably be induced to move, and that some backward regions lack the potential for eventually self-sustaining growth in employment or that some developed but distressed regions must inevitably shrink in size in order to adjust to new economic conditions, the strategy implications are the opposite of those just described. We conclude that many of the unemployed people will best be served by moving to some area with better opportunities, and we draw a sharp distinction between their welfare (people prosperity) and place prosperity. Assistance logically takes the form of improving the employability and mobility of the people affected, facilitating their relocation, and promoting employment opportunities in the areas of greatest potential. Thus the elements of this position are people prosperity, stimulation of development on the basis of growth potential, and stress on the upgrading of human resources. Job creation does not have to be stimulated on a diffused basis in a large number of individual areas, since people are prepared to move to one of a smaller number of growth centers.

Which of the foregoing two positions is the more correct? Clearly, neither is wholly right or wrong, since both people and employment activities are partially footloose. A few observations are in order, however.

First, certain emotions and prejudices seem, on balance, to impart bias toward the view first mentioned (namely, that jobs must move to people). Because of local pride as well as vested interest in their community or region, most regional spokesmen are reluctant to admit that their region lacks development potential or to see its population decline. As we noted earlier, most of the active and articulate spokesmen and leaders in regional development are those who do have a vested economic interest there in the form of large property ownership, a business depending on local markets, or a political position whose importance and perquisites depend to some extent on the region’s size and growth. Quite naturally, they are ready to invoke ethical and cultural arguments in support of their economic interests and loyalties.

It is an article of faith among many that people should not have to move in order to better themselves, any more than they should have to change their religion, political affiliation, or skin color. A presidential advisory commission in 1967 endorsed "a national policy designed to give residents of rural America equal opportunity with all other citizens. This must include access to jobs, medical care, housing, education, welfare, and all other public services, without regard to race, religion, or place of residence." ²⁸

Reinforcing this bias is a general tendency to overrate the footlooseness of activities with which one is not directly familiar. In particular, the complex and subtle economies of agglomeration that favor major urban areas as locations are not well understood. Moreover, the consideration of efficient interregional allocation of resources and output, from the standpoint of national welfare, has few spokespersons. That faceless individual, the consumer and taxpayer, is here again the forgotten person.

In view of this considerable bias, it is not surprising that official policies and public statements have generally soft-pedaled migration as an instrument of regional policy, have paid a great deal of deference to the place prosperity strategy and the criterion of need, and have favored spreading assistance among an increasingly large number of claimant areas rather than concentrating it.

How mobile are people in areas of high unemployment, and can their mobility be expected to increase? A number of excellent studies have addressed themselves to these questions.²⁹ First, it appears that unemployed people (regardless of area) are more likely to want to migrate than are employed people of the same occupational or age group. But these desires tend to be frustrated in the case of the less educated, the less skilled, and the black. John Lansing and Eva Mueller conclude that:

unemployment constitutes a "push" which leads people to move if they are young, well-educated and trained, or live in a small town. In the absence of such characteristics, unemployment is highly unlikely to overcome the reluctance to move, unless the unemployment is prolonged, the income loss substantial, and the family has no alternative local source of support.³⁰

Thus the labor force groups most prone to unemployment are also the least mobile (quite naturally, because they’ have the least to offer in relation to labor demands and the least likelihood of finding a job if they do move). Out-migration is highly selective in favor of the better trained and more educated. This has two serious implications. First, even assuming continuous prosperity, we cannot presently count on migration alone to solve all the problems of distressed areas by draining away their unemployed. Second, such migration from distressed areas as does occur results in a lowered "quality mix" of the labor supply of those areas, which may further handicap them in any competition for new employers.

But if migration is inadequate, the remedy is not to discourage it, as some would propose. It is quite possible and certainly more appropriate to upgrade the less productive and less mobile groups so that they will be better able to migrate and also will be more attractive to potential employers wherever they are. One of the great virtues of a strategy of human resources development, improved job information, and placement services is this double-action impact. It helps people move to jobs and helps jobs move to people. The danger in practice is that part of the benefit may be thrown away by misguided efforts to restrict migration— for example, by training people only for the kinds of jobs existing in their home areas, or by pension plans, union restrictions, and relief eligibility rules³¹ that discriminate against newcomers in areas of in-migration.

The long-term prospects—or at least the possibilities—seem good for some continued increase in the mobility of the disadvantaged groups in labor surplus areas; this should diminish migration selectivity and allow migration to contribute more effectively to regional adjustment.

The mobility of employment locations is the other important aspect relating to the issue of bringing jobs to people or the reverse. It is commonly said that manufacturing industries have become much freer in their choice of locations than they were in the age of coal and steam, and this is almost certainly true as among regions. It is not at all obvious, however, that employers are becoming increasingly indifferent about where they locate. There have been substantial population shifts in the last decade or so, and these have been mirrored to some extent by changing patterns of growth in employment.³²  For many companies, smaller cities, towns, and unincorporated places are becoming increasingly attractive location alternatives. For others, the nation’s large metropolitan areas continue to offer important advantages. In either case, the decision to locate is not a matter of whim and fancy but is guided by economic incentives.

In any event, there seems to be ample evidence that an attempt to solve problems of regional employment by bringing new industry to every community or labor market area would be wasteful and futile. Henry Ford I in the 1920s, and many others before and after, have thought it possible and desirable that industrial employment be diffused to every small town and village, and the first Indian Five-Year Plans after independence put substantial reliance on developing small-scale village industries. In no country, however, has such an attempt really succeeded.

On the contrary, shifts of population and employment to major urban areas and out of small towns and the countryside reflect in part the growing importance of tertiary activities, the declining importance of agriculture, the improvement of long-distance communication and people transport,³³  larger-scale production and management units, demand for urban-type amenities, and proliferation of the external economies of agglomeration and urbanization.

Counter movements to smaller communities are similarly selective. Manufacturing activity may respond to the existence of a skilled work force that is particularly well suited to the production of high-technology components, and service industry growth may follow the population movements of retired persons to amenity-rich rural areas, but not all places—metropolitan or nonmetropolitan—share these characteristics equally. For example, David L. Brown reports that some 20 percent of all nonmetropolitan counties continue to experience out-migration and population decline in the face of the population turnaround of the 1970s.³⁴

12.5.3 Some Conclusions

Where does all this leave us in terms of the basic strategy issues for regional development assistance? The points raised thus far suggest these conclusions:

1. Migration can, does, and should play a substantial role in effecting desirable regional adjustments. Its effectiveness tends to grow and can be greatly enhanced by programs of education, training, retraining, equal opportunity, open entry,³⁵  job information, and placement services especially directed at the least employable and least mobile manpower groups in areas of labor surplus. Programs more explicitly directed at the encouragement of migration can also play a substantial role.³⁶

2. Employment is not fully footloose: There are important differences in the development possibilities of different areas. It would not be feasible to bring employment (except of the work relief type) to each and every labor market area.

3. Accordingly, place prosperity is an inadequate and misleading goal; development assistance should be allocated on the basis of the needs of people and the development potential of areas; such assistance should be at least to some extent focused on particularly promising locations; and human resources programs of the type outlined in (1) above should play a major role.

4. Strong political pressure is to be expected in the direction of the use of local distress as a priority guide, the discouragement of emigration, and the diffusion of assistance to more and more areas.

12.6 THE ROLE OF GROWTH CENTERS

One of the four basic issues of regional development assistance strategy concerns the focusing of such assistance upon a relatively small number of selected growth centers,³⁷  at which there exist or can easily be created the necessary conditions for expanding employment opportunity and, especially, the public infrastructure and the external economies that most activities require. Such growth centers are then expected to attract commuters and migrants from surrounding areas of labor surplus, and at the same time to stimulate secondary growth of employment in some of those areas.

12.6.1 Applicability of the Growth-Center Strategy to Different Types of Problem Areas

The problem of choosing growth centers arises only in certain of the problem areas characterized in section 12.3. There has been a tendency, in assistance programs, to lump together indiscriminately the backward areas and the developed but distressed areas.

The two types of areas do share, of course, certain symptoms of maladjustment. Both suffer essentially from obsolescence of the bases for their former economic viability; both need help in making a structural shift to a new base in response to changes that have occurred in demand, resources availability, and competition from other areas. For both, a successful transition calls for modernizing human and capital resources and infrastructure (including institutions and attitudes) so that they can effectively grasp new opportunities provided by technological and economic change and thus become more resilient, self-reliant, and generative.

But at this point the similarity ends. With respect to needs for education, the two kinds of areas are likely to differ substantially. The population of a distressed developed area may show no particular deficiencies in all-round literacy and capability for productive industrial or tertiary employment. Internal and external transport and communication facilities in such an area are also likely to be adequate or more than adequate. There are substantial local resources of capital and at least some relevant industrial know-how. The basic elements of growth centers are already there, and the problem is essentially one of modernization—reorienting the local labor force, business community, infrastructure, and public sector toward the opportunities of today and tomorrow.

By contrast, for truly backward areas with little industrialization or urbanization, the necessity of finding or creating specific growth centers is of major concern. It is primarily to this kind of region that we refer here.

12.6.2 Justification for Focusing Employment Stimulus in Growth Centers

The next question that concerns us is the role the growth center is supposed to play vis-à-vis the surrounding area. On both economic and political grounds, it is vital to have an acceptable answer to this question, if only to justify the denying of direct aid to places that are not growth centers. Justification is needed because these other places cannot be expected to like being left out of the distribution of largess, and because the growth centers are likely to be relatively well-off and growing places and thus apparently the least in need of any help. "Unto every one which hath shall be given" is scarcely a policy to evoke the enthusiastic support of a "hath-not" area.

Two elements appear in the case usually made for the growth-center strategy. The first argument stresses availability of infrastructure and the external economies of urban size as prerequisites for competitive survival in a modern economy. Concentration of public investment at growth centers is justified on the ground that those are the only locations where adequate public services can be provided at reasonable cost and where there is a prospect that prosperity and growth can eventually be self-sustaining without permanent subsidy.

This basis of strategy was clearly involved, for example, in a project proposed by the Québec provincial government in late 1969 for the Gaspé Peninsula, where eleven backwoods villages were slated to be wiped off the map. The residents would be given cash incentives to relocate in larger coastal towns where schools, hospitals, and vocational training centers could be made available. The project was described as merely the initial experimental stage in a larger development program for the backward rural areas of the province.³⁸

Were this the only rationale for the growth-center approach, it would imply that the peripheral backward areas outside of the centers have no prospects of survival except as charity cases, and that they should be vacated as fast as is humanely possible. But there is a second argument in this case; namely, that some of the effects of economic improvement initiated in growth centers will spread out to their less developed hinterlands or zones of influence. This implies that the best way to help these hinterland areas may be not by either uprooting or direct assistance but indirectly through promoting the progress of accessible growth centers. Let us see how this spread effect may be expected to work.

There was mention in Chapter 11 (see Section 11.7) of the manifold ways in which an urban center can provide a focal point of leadership in the development of its region. All the considerations mentioned are relevant to the growth-center strategy, but we still do not know a great deal about how to measure or control the effects in question. Most of our quantitative knowledge is in terms of the two familiar frameworks of central-place and input-output analysis. Each of these approaches is helpful only to a limited extent in articulating the impact of a growth center on its zone of influence.

The central-place model is designed, in fact, to describe essentially the inverse relationship; namely, the dependence of the urban center on demand in its tributary area. Central-place analysis is concerned only with a limited set of consumer-serving activities that are stringently market-oriented. The spatial distribution of consumer demand is taken as given, and it determines the extent to which various orders of central places can develop appropriate ranges of consumer-serving activities.

Despite the fact that the roles of the central place and growth center are so different, the analysis of a region’s system of central places may be important. First, the central-place analysis will indicate something about minimum size constraints. It can establish that cities or towns below some specified population are unlikely to contain certain trade and service activities that may play an essential role in the operations of a growth center. Second, the tributary trading area of a central place, being based largely on the feasible range of frequent travel, may be a rough indicator of the zone of influence that place would have as a growth center. Third, the central-place hierarchy can serve as a mechanism by which innovations are transmitted interregionally, and growth centers may be important links in that network.³⁹

The structure of the central-place hierarchy in a country can also affect the success of growth-center strategies. Typically, less developed nations lack an integrated system of cities; as mentioned previously, they are characterized by a chief (or primal) city and many small villages, but cities of intermediate size are underrepresented in the urban hierarchy. The locational influence of agglomeration economies in the dominant city may be difficult to overcome under these circumstances. It would be necessary to concentrate the resources available for development programs in a very small number of designated centers if producers in these places were to compete effectively in the national market. In this context, growth centers would also serve to bring much needed public and private services to backward regions, reducing the attractiveness of the primal city for rural residents.⁴⁰

The relations described by the input-output model are more directly relevant to the role of a growth center, particularly if we think of a model embracing as separate subregions the growth center and the zone of influence. In its usual application, the input-output model traces direct, indirect, and induced impacts of some initial change via backward linkage, and this is of course one of the mechanisms by which a growth center can stimulate its tributary zone. For example, manufacturing and other exporting activities in the growth center will purchase local materials and services, some of them from the zone of influence. A food-processing plant illustrates the direct effect. By its presence in the growth center, it provides a market for farmers in a surrounding agricultural area. In Chapter 11 we explored the nature and measurement of the subsequent indirect effects (through local purchases by business firms) and induced effects (through local purchases by households). For as much of the zone of influence as constitutes the commuting field of the growth center, the most obvious impact of growth at the center on the surrounding area is likely to be the direct, indirect, and induced demand for labor.⁴¹

In principle, as was suggested in Chapter 11, input-output analysis can provide insights relevant to the evaluation of forward linkages. But input-output analysis is severely constrained in the extent to which it can express the role of growth centers because, for operational reasons, it ignores the scale economies and the external economies of agglomeration that are basic to the whole growth-center strategy. Nor does the input-output approach, as developed so far, take into account growth-initiating factors—such as the supply of capital, enterprise, and specific public services, or the progressive improvement of productivity through education, health, training, and informational services.

It is clear that growth centers exert their influence in many ways that elude the usual quantitative models and systems of accounts. In particular, there is a recognized need for more adequate techniques for dealing with those growth-center effects that operate through supply rather than through demand.

It is difficult in principle to make a meaningful distinction between the forward-linkage effects and the external-economies effects of growth-center development; in both cases an activity initially established in the center provides cheaper and more accessible inputs that make possible the nearby establishment or expansion of other activities dependent on access to such inputs. Generally, we seem to prefer to speak of external economies when the initially established activity is of a so-called threshold type, normally associated (because of scale economies) with a certain minimum size of urban or industrial concentration, and when it provides products or service inputs to a wide variety of other activities in the same locality. We are more likely to refer simply to a forward linkage when those conditions do not hold and when the initially established activity supplies inputs to just one or a few activities locationally oriented to sources of that input. But both cases involve a similar principle of input orientation or forward linkage. Finally, terms such as "infrastructure" or "social overhead" generally denote services supplied by the public sector or by public utilities—such as schools, hospitals, water supply, and communications.

The transmission of growth effects outward from a growth center via forward linkages involves only those kinds of outputs that can be transferred from the center to the people in the tributary region. This would not ordinarily include fire protection, elementary schools, or garbage disposal. It does, however, include a wide variety of public services (technical schools, colleges, research libraries, and hospitals) and a similarly wide variety of business services (commercial research and testing laboratories, banks, data-processing centers, and so on).

12.6.3 Size and Number of Growth Centers

Adoption of the strategy of growth centers already implies a substantial degree of geographical focusing of development assistance. But should the growth centers be few and large, or numerous and small?

There are a number of possible approaches to this question. For example, we can examine the past growth records of urban areas of various sizes to see whether size seems to affect growth potential in any important way. Here we would have conflicting evidence. On the basis of the rapid growth of metropolitan areas throughout the 1950s and 1960s, many researchers surmised that the realization of agglomeration economies was both necessary and sufficient for sustained regional growth. It was argued that a population of 250,000 could be regarded as a threshold in the development process; after that size had been attained these economies could be expected to ensure continued growth.⁴²  However, the data on population growth presented in Chapter 11 (see Table 11-6) show that smaller places participated fully in the rapid growth of the regions of the South and West, and that metropolitan areas in the Northeast actually lost population, during the 1970s. We may find that agglomeration economies are important in nonmetropolitan growth, just as they are in metropolitan growth; but there is scant evidence on this to date. In any event, the proliferation of small growth centers would be risky and expensive. It is easy to see that sound evidence on the role of agglomeration economies in the context of metropolitan area decline and nonmetropolitan area growth is sorely needed.

Another approach is to try to estimate the costs of providing basic public services or infrastructure in cities of various sizes, to see whether we can identify urban-size economies and an optimum size or a minimum efficient size of city with regard to such costs. Here we get some limited guidance, to the effect that middle-sized cities (say from 200,000 to 1 million population) tend to have lower unit public service costs than smaller places, and (with some what less certainty) lower costs than the still larger places (see Section 12.3.3). But some difficult problems are involved in making legitimate comparisons of such costs between one city and another. (Just how, for example, do we measure the cost per unit of output of a police force or a park system?) More important still, the costs and efficiency of public services (even if we could measure them accurately) would be only one element in the comparative social costs and effectiveness of different sizes of cities viewed as agents for the development of surrounding zones of influence. There is no reason to suppose that the optimum size of growth center coincides with the minimum cost size of city from the limited standpoint of measured public service costs. Thus all that we really get out of this approach is a warning that when we dip down into, say, the five-figure population range, the potential growth center is increasingly likely to be handicapped.

Still another approach leans on central-place theory and data, and attempts to define a viable growth center in terms of the range of central-place activities represented there. Leaders in developing this approach were Karl Fox and Brian Berry. The range of the center’s influence as a purveyor of consumer goods and services and the range of its influence as an employer of labor are tied together in the concept of a "fundamental community" or functional economic area demarcated on the basis of both commuting and shopping distances and having a sufficiently full line of central-place activities to be relatively self-contained.⁴³

If the area radius is assumed to be large and if we are willing to accept quite small cities as nuclei of functional economic areas, the network of such areas can be spread out to cover the bulk of the population of the United States. For example, Berry constructed a set of such areas that included in their boundaries 96 percent of the 1960 population. The radius in this case was based on one hour’s driving time as the limiting factor, with an assumed average speed of about 50 miles an hour. Many of the central cities were well below 50,000 population.⁴⁴

In the Fox-Berry conception, it is not the size of the central city or growth center that matters but the population of the whole commutation and trading area including it. Such an area is regarded as constituting a single community, and Fox has suggested that something like 250,000 might constitute a viable size for self-sustaining growth.

It does not yet appear established, however, that a population of 250,000 spread over 7,900 square miles (the area of a circle of 50 miles radius) and lacking any city of more than 25,000 population would have the same growth potential as a metropolitan area of 250,000, which might be expected to contain a single county with an area of just a few hundred square miles and a central city of 60,000 to 150,000 population. The realization of external economies depends often on the density of population (proximity). This is true for business establishments; but as mentioned earlier, it applies also to the provision of public services. Thus, the spatial distribution of population undoubtedly matters.

All this discussion of growth-center strategy suggests that a major policy problem is how to avoid yielding to the pressures for too much proliferation of growth centers and spatial diffusion of development investment.

12.6.4 Migration to Growth Centers

The growth-center strategy is sometimes presented as an alternative to migration from backward rural areas and small towns. Nevertheless, it would appear that migration does and should play an important role in a successful growth-center strategy.

First, "commuting range" is a somewhat elastic concept. The 50 miles suggested by Fox is certainly feasible with automobiles and good highways; but most people prefer not to commute that far if they can avoid it. Growth of employment opportunity in a growth center normally will attract from such distances people who initially commute but eventually move closer. Some local inward migration within the zone of influence of a growth center is, then, a part of the development sequence.

Second, it would be entirely unrealistic to expect a regional development strategy to eliminate incentives and need for migration among zones of influence of various urban centers. No development plan can or should aspire to make the growth of employment opportunity in each regional or labor market area exactly match the natural increase rate of the working-age population.

Finally, our consideration of the size requirements for a viable growth center suggests that in many poorer and less developed regions substantial areas will lie outside of even a 50-mile range of any urban center of sufficient size and promise to merit growth-center status. Though no one would propose the blanket evacuation of all such areas, it seems clear that most or all of their natural increase of population, and perhaps also some of their existing population, will need to move out in order to find adequate opportunity for self-supporting employment.

Some researchers have criticized growth-center strategies precisely because of these polarizing effects; they argue that the selective nature of migration constitutes an important negative or backwash effect of this policy on outside areas.⁴⁵ While the effects of migration can have serious consequences for areas losing population, it is important to recognize that this is but one phase of a development process that may take many years to complete. The most apparent and immediate effects of the growth center may well be migration and a subsequent depletion of human resources from some areas. However, the beneficial consequences of this policy may take longer to be realized.⁴⁶

An assumption implicit in the growth-center strategy is that people in backward regions will migrate more readily to a growth center in their own region than they will go to places outside that region. Distance is assumed to be an important determinant of migration flow.

Proximity does indeed seem to encourage migration—thus bearing out one of Ravenstein’s Laws and the deductions of theorists. The actual costs of moving are probably less important in this connection than the "social distance" involved in moving to an area with very different characteristics and climate and in which there is a smaller probability of knowing someone (see the beaten-path principle cited in Chapter 10).

But the big hurdle to be overcome in inducing migration from backward areas to employment centers involves the initial decision to move at all. The social distance from any farm or village to a sizable city is enormously greater than that separating different rural areas or different urban places of similar size, and many people in backward areas suffer special disabilities of lack of education, training, and information. The evidence does suggest that such people will probably move more readily to a growth center within, say, a hundred miles from home than they would to an entirely different part of the country; therefore, the creation of more jobs in such growth centers will help them in getting employed. But clearly the strategy must also involve measures to improve mobility and employability as such and to facilitate entry to productive employment at the growth center.⁴⁷

The implication of focusing attention on employment in rather few growth centers, all of substantial size, is that many or perhaps most such centers will lie outside the boundaries of regions demarcated on the basis of such indices as high unemployment, low income, or slow growth. Measures aimed at stimulating employment by improving infrastructure and inducing private investment may thus be most effective when applied in places already relatively well-off, active, and prosperous; while measures applied to less urbanized and poorer areas may be confined largely to human resources development and to income supplements for people who, because of age or disability, cannot be expected to solve their problems by migrating.

12.7 ASPECTS OF UNITED STATES REGIONAL DEVELOPMENT PROGRAMS

The foregoing discussion has disclosed the main policy issues involved, in national efforts to assist regional development as a means of improving people’s welfare; it has suggested solutions with which the reader may or may not wholly agree. It has likewise noted some of the administrative and political difficulties complicating strategy decisions and their implementation.

Let us now see how these problems have been handled in the major programs of regional development assistance in the United States. We shall not try to deal with the strategies and programs (many of them quite similar) that have been developed in other countries. Nor shall we go into much detail regarding the American experience, since programs focusing on the development problems of specific regions have been phased out almost completely in recent years. Rather, we shall concentrate on exposing the character of U.S. regional policy generally by discussing the emphasis and goals that can be discerned.

National concern with regional development was sparked during the presidential campaign of 1960. Once elected, John F. Kennedy fulfilled a campaign promise by introducing legislation for a comprehensive development program aimed at depressed areas, in particular Appalachia (defined officially as including all of West Virginia and parts of 12 other states, stretching from New York, Pennsylvania, and Ohio in the North to Alabama and Georgia in the South). This initiative culminated in the establishment of the Appalachian Regional Commission (ARC) and the Economic Development Administration (EDA). The latter was set up in the U.S. Department of Commerce by the Public Works and Economic Development Act of 1965 to replace a predecessor agency, the Area Redevelopment Administration (ARA). The discussion to follow will focus first on ARA and EDA before turning to regional commissions such as ARC.

In recent years, there has been a change in emphasis from concern with the problems of designated regions to concern with urban areas facing a common set of problems. In Chapter 13, some aspects of development policy will be discussed in the context of problems associated with fiscal distress in central cities.

12.7.1 ARA and EDA

Both ARA and EDA were established to enhance employment opportunity in specific areas by making aid available that might encourage the expansion of private enterprise. The essential difference between the two is that the earlier ARA emphasized direct assistance to firms in the form of grants or loans if they established plants in designated areas. Although the business development loans and loan guarantees that characterized ARA continued as EDA programs, EDA funds have been concentrated on assistance to local public authorities to help build or improve public service facilities (such as sewer and water systems) that would provide the infrastructure necessary for the development of commercial activity.

The public works character of EDA expenditures is brought out clearly in Table 12-2. There we find that cumulative expenditures by EDA on approved projects totaled just under $9.6 billion as of March 1978. Of that amount, 35.9 percent, or roughly $3.4 billion, were committed to development programs as a whole, and 65.9 percent of these funds (just under $2.3 billion) were for public works. Other development funds were used to aid areas in planning and analysis through technical assistance in kind or through grants to support such work. Additionally, slightly more than $6 billion went to other "nondevelopment" programs, primarily a contracyclical local public works program initiated as a result of the serious recession of the mid-1970s.

This emphasis on public works projects is undoubtedly the product of diverse political considerations. Programs of this sort are especially attractive to members of Congress who are able to point to tangible evidence of their ability to bring federal dollars to their districts, hoping to enhance their chances for reelection. They also represent a "low-risk" policy; in this respect, the sentiments of many observers have been expressed succinctly by William H. Miernyk: "It is much safer to invest in public works, where a complete failure is difficult to define."⁴⁸  As Miernyk points out, even if the public works projects fail to stimulate private investment, the communities involved end up with a new sewer system, a better bridge, or some other public facility; whereas if money is spent to plan and construct the facilities for an industrial park, and it goes unused, the failure is there for all to see.

Areas eligible for development programs were defined by EDA at three different levels of size. For projects of only local significance, the unit was the "redevelopment area," which could be as small as a county, a city, an Indian reservation, or in certain cases even smaller.

Eligibility for assistance could be established on the basis of any of a number of criteria. As of September 30, 1981, there were 2654 areas (encompassing over 80 percent of the United States)⁴⁹ that qualified for assistance under various criteria. The criteria, and the number of areas that qualified under each, are given in Table 12-3.

For both EDA and ARA, the policies of place prosperity, worst-first, and reliance on stimuli to employment-creating investment appear dominant. One of the reasons for the new 1965 legislation, however, was dissatisfaction with the exclusive emphasis on the approach under which ARA operated. An important feature of the EDA program is the creation of a set of larger units, known as economic development districts. Each such district must contain at least two redevelopment areas plus at least one "economic development center" (growth center), and those centers are eligible for assistance of the types already described. Local initiative to form development districts is stimulated by a provision for extra funding for redevelopment areas that are part of development districts. The economic development center must have "sufficient size and potential to foster the economic growth activities necessary to alleviate the distress of the redevelopment areas within the district," but it must not have a population of more than 250,000.

The stated purpose of this application of the growth-center concept is "that economic development projects of broader geographical significance may be planned⁵⁰  and carried out." Although size and potential are recognized as criteria for aid to the growth centers, and the latter are recognized as useful in helping the distressed areas, the act makes no mention of the possibility that people in the redevelopment areas might be helped by migrating to the growth centers. Our earlier discussion suggested that such migration is probably the principal way in which the growth centers can help, and that quite different strategies are appropriate in growth centers and in distressed outlying areas respectively. But EDA’s mandate appears to assume that the same kinds of assistance are appropriate in both places.

Another point to be noted is that there is no specific minimum population size for growth centers, though there is a maximum of 250,000 (corresponding to what has been suggested by some regional economists as a minimum for self-sustaining growth!). The standards provided in the law do not provide much resistance to the predictable local pressures for designation of an ever-increasing number of small development districts and centers, since a combination as small as just two poor counties and a town could be designated as a district.⁵¹

Since 1981, EDA has barely survived federal budget cuts. Its operations continued, although at a greatly reduced level, through 1984 as a result of funds made available by congressional resolutions, even though EDA was omitted from the budgets submitted by President Reagan. Prospects for the continuation of EDA are uncertain at best.

12.7.2 The Regional Commissions

Title V of the same Public Works and Economic Development Act of 1965 provided for designation of still larger areas, called economic development regions, extending into two or more states. Each such region had a regional commission made up of a representative of each of the states involved plus a federal representative with veto power. The prototype was the Appalachian Regional Commission, established under separate legislation in 1965 but with similar purposes and powers. As of 1983, only the Appalachian Regional Commission continued to receive federal support; however, it was already committed to a five-year finish-up program at that time. For our purposes, we can consider them all together, including Appalachia.

Economic development regions were defined on the basis of

1. High unemployment

2. Low income

3. Low levels of "housing, health, and educational facilities"

4. Dominance of the regional economy by "only one or two industries, which are in a state of long-term decline"

5. Substantial out-migration of labor, capital, or both

6. Low growth rate of aggregate output

7. Adverse effects from changing industrial technology or changes in national defense facilities or production

ARC and the Title V Regional Commissions were primarily designed to secure interstate cooperation and a broader perspective for development planning and action within areas much larger than development districts. They were expected to produce plans for a coordinated attack on the economic problems of their respective regions through all kinds of existing and proposed federal and state programs. They were, of course, in competition with one another for federal assistance, though some respect for the general national interest was expected to be introduced by the federal cochairman of each commission. Each regional commission had advisory functions regarding the initiation and coordination of economic development districts.⁵²

By the end of 1972, the economic development regions shown in Figure 12-1 had been established. Including Appalachia, they encompassed part or all of 39 states and included such major metropolises as Boston, Pittsburgh, New Orleans, St. Louis, Kansas City, Seattle, and Portland. Most of these regions had been showing relatively slow growth for some time and have extensive unindustrialized and even backward areas.

The inclusion of New England may seem surprising. New England is the patriarch of American regions in terms of industrial and urban development, has an income level comparing favorably with the national average, and does not have especially high unemployment. If it is a "problem area" at all, it cannot be rated as such on the basis of underdevelopment or poverty as can the others.⁵³  The inclusion of New England probably reflects instead a positive factor. The region has been ahead of others in achieving a sense of common regional interest and developing effective interstate cooperation. This early start reflects the facts that the challenge of industrial stagnation came early in New England, reaching almost crisis proportions in the 1920s with the loss of the textile and other industries, and that the New England states are very small in size compared with those in other parts of the country.

The establishment of New England as an economic development region suggests, in fact, that it would be appropriate to carve up the whole country into development regions instead of using them as special devices for recognizable sick areas of wide extent. An initiative of this sort was undertaken in the late 1970s, when "wall-to-wall" commissions were established. While they did not survive the fiscal austerity characteristic of domestic programs in the 1980s, this extension of commissions modeled after those of Title V was at least tacit recognition that a homogeneous region defined in terms of backwardness and distress is not the proper unit for constructive and efficient development policy.

Appalachia, for example, is not a region at all in the sense of an area with strong internal linkages. Appalachians in Pennsylvania or West Virginia may resemble Appalachians in Alabama with respect to income level, education and training, style of life, and attitudes; but they are not linked to them by any significant flows of trade or migration. A conspicuous characteristic of Appalachia, in fact, is the lack of facilities for internal movement. Much more meaningfully, Appalachia should be regarded as a succession of hinterlands to various major centers located mainly outside the region as officially defined: a row of back yards, as it were. We are led to the view that

a good part of Appalachia’s development effort should be concentrated outside the region, and . . . the region itself should be restructured and, as it were, apportioned among the metropolitan regions on its perimeter.⁵⁴

Similar statements could be made about some of the other economic development regions as well.

It is true that development assistance in Appalachia has heavily emphasized roads, with the avowed intention of "opening up" the region to the outside world and to such cities as it contains. But evidence is lacking of any effort to encourage out-migration from this or the other economic development regions. On the contrary, loss of population is stipulated as one of the criteria of eligibility for development assistance; and by implication at least, as one of the conditions to be corrected.

Economic development regions can serve useful and constructive ends. To be effective, however, they must be structured around one or more major growth centers with demonstrably high potential. Such a region would have the internal cohesion that is missing in an area such as Appalachia. Additionally, it would have the political, administrative, and technical resources necessary to plan and implement effective development strategies.

The regional problems that were of so much concern in the 1960s and early 1970s have not gone away; their character has simply changed. The nonmetropolitan growth that has been characteristic of recent population trends has reduced the pressure for policy makers to focus their attention on "backward" areas, but many of these remain in distress despite the nonmetropolitan resurgence. Additionally, the slow growth of metropolitan areas, particularly in the Northeast and upper Midwest, is likely to persist for some time; transition of the economic base in such cities as Youngstown, Akron, and Buffalo to new growth industries is not going to happen overnight.

The coordination of federal and regional efforts can facilitate the type of change that must take place in distressed areas (metropolitan or nonmetropolitan) and keep it from being jeopardized by self-serving local interests or potentially destructive interregional rivalry. A system of regional commissions characterized by a combination of intraregional interdependence, greater consciousness of a common regional interest, and financial and technical strength would serve this end; and it could help also to check the trend toward ever-increasing dependence on federal initiatives, decisions, and specific subsidies. The federal role could be less paternal and philanthropic, and could consist mainly of maintenance of high overall levels of demand, the provision of information and aggregate national development guidelines, and the design of programs to improve the quality and mobility of human resources. A system of this sort would be a step toward the decentralization of federal authority which would ensure that the regions assuming new responsibilities meet at least some of the necessary criteria for success.

12.8 SUMMARY

The formulation of public policies on location and regional development was stimulated in the second quarter of this century by continuing interregional disparities in income and economic opportunity, by the increasing role of the national government in financing and providing regional services, by disenchantment with population increases as an objective and with competitive regional subsidization of growth, by the dilution of provincialism, and by changes in factors affecting location.

National government programs to maintain high employment levels and to improve manpower quality and mobility are warranted on grounds of both equity and efficient allocation of resources among and within regions. An important distinction exists, however, between the objective of "place prosperity," or economic assistance to regions as such, and the really fundamental goal of "people prosperity."

"Problem regions" are of several different types, including (1) backward areas halted at the threshold of self-sustaining development; (2) already developed areas with arrested growth due to loss of competitive advantage in their basic activities or obsolescence in those activities as such, with accompanying loss of ability to substitute new kinds of activities; and (3) areas of excessive growth or excessive concentration.

Public policy can influence regional structure and development through many measures, which include upgrading manpower quality and factor mobility, maintaining a high national employment level, subsidizing or restricting investment, controlling transfer rates and services, allocating public purchases and investments among regions, supporting research and development, and assisting in the provision of local or regional infrastructure.

Four main issues arise in connection with public policy toward regions: (1) degree of reliance on the place prosperity criterion, (2) allocation of regional assistance as charity or as investment, (3) focusing of assistance in growth centers as contrasted with wide dispersion, and (4) choice among available devices for influencing development.

Throughout the 1960s and 1970s regional development policy in the United States favored "backward" regions. More recently, there has been a shift in emphasis toward distressed urban areas. Certain built-in political and economic biases, evident both in this country and abroad, have led to overemphasis on place prosperity, overproliferation of areas receiving public developmental assistance, and underrating of the potential role of migration and mobility enhancement.

SELECTED READINGS

John Friedmann and William Alonso, Regional Policy: Readings in Theory and Applications (Cambridge, Mass.: MIT Press, 1975).

Gordon C. Cameron, "Growth Areas, Growth Centres and Regional Conversion," Scottish Journal of Political Economy, 17, 1 (February 1970), 19-38.

Ira S. Lowry, "Population Policy, Welfare, and Regional Development," in Mark Perlman, Charles Leven, and Benjamin Chinitz (eds.), Spatial, Regional, and Population Economics (New York: Gordon and Breach, 1972), pp. 233-261.

Niles M. Hansen (ed.), Public Policy and Regional Economic Development (Cambridge, Mass.: Ballinger, 1974).

Harry W. Richardson, Regional Economics (Urbana: University of Illinois Press, 1978), Chapters 7, 9, and 10.

E. A. G. Robinson (ed.), Backward Areas in Advanced Countries (London: Macmillan, 1969).

Norbert Vanhove and Leo H. Klassen, Regional Policy, A European Approach (Montclair, N.J.: Allenheld, Osmun, 1980).

APPENDIX 12-1

The Shift-Share Analysis of Components of Regional Activity Growth

See section 12.3.2 and 12.3.5

The overall growth rate of a region’s activity (as measured, say, by total employment or total value added) is, of course, a weighted average of the growth rates of the separate sectors or activities making up the region’s economy. If the region’s growth rate is compared with that of some other area (for example, the entire nation), it is possible to "explain" the difference in growth rates statistically in terms of two components, which for convenience can be styled "mix" and "competitive." Quantitative analysis of comparative regional growth rates along these lines is sometimes referred to as the "shift-share" approach.⁵⁵

An example of a regional growth differential arising exclusively from mix would be the case of a region in which each activity grows at exactly the same rate as in the nation as a whole. In other words, the region’s share of the national total for each industry remains unchanged over the time interval in question, but the national growth rates for some activities are higher than those for others. If a region contains mainly fast-growing activities and relatively few of the slow-growing activities, it can be said to have a "favorable growth mix" of activities, and its overall percentage growth rate will exceed that of the nation. On the other hand, if slow-growing industries are more than proportionally represented in the region’s mix, the region’s overall growth rate will be slower than the national growth rate. This is an example of the pure mix effect.

We can evaluate the competitive component by imagining the case of a region that has exactly the same mix of activities, as does the nation:

Its percentage share of the national total is the same for all activities. This region will have an overall growth rate higher than that of the nation if it increases its shares (that is, if most activities grow faster in the region than in the nation). Such a case represents the competitive component in isolation.

In any real situation, of course, it is nearly certain that the relative growth rates of region and nation will show the effects of some combination of mix and competitive components. Either effect, or the net result, can be either positive or negative for the region.

A drastically simplified numerical example will serve to show the way in which shift-share analysis determines the effect to be imputed to each component. Let us take the following Census data on manufacturing employment (in thousands):⁵⁶

From these data, we see that manufacturing employment in the United States increased by 5.79 percent (from 15,800 to 16,715 thousand) in the five-year interval. If manufacturing employment in Pennsylvania had registered this same rate of increase, it would have risen from 1,331 to 1,408 thousand persons. Actually, the 1963 employment in Pennsylvania was only 1,320 thousand, so there is a total difference of -88 thousand to be explained.

To evaluate the mix component, we can eliminate the competitive one by assuming that each of the two kinds of industries grew at the same rate in Pennsylvania as in the nation (that is, durable goods industries 9.61 percent and nondurables 2.18 percent). Had those sectoral growth rates applied in Pennsylvania, the state’s manufacturing employment in 1963 would have been 718 X 1.0961=787 thousand in durables and 613 X 1.0218=626 thousand in nondurables, or a total of 1,413 thousand. So the mix effect operating in the absence of any competitive effect would have raised Pennsylvania’s manufacturing employment to 1,413 thousand, or 5 thousand more than what would have been achieved by simply keeping pace with national growth (i.e., 1,413 — 1,408=5). We can express this result by saying that Pennsylvania had a favorable growth mix compared to the nation (in this case, meaning a higher proportion of durable goods industries in its mix), and the 5 thousand figure is a measure of that advantage, or the mix component of growth.

But as noted earlier, Pennsylvania’s actual growth fell 88 thousand short of what would have been achieved by keeping pace with the nation. The competitive component, then, must be —88 — 5=—93 thousand. This figure is a measure of the result of the fact that Pennsylvania’s share of the national total dropped in both durable and nondurable goods industries; that is, Pennsylvania industries lost out to that extent in competitive position vis-à-vis the rest of the country.

The results of this dissection of growth components are diagrammed in Figure 12-1-1. All the figures are expressed in absolute terms (thousands of employees), since they are additive. In comparing the mix and competitive shifts of different regions, however, it might sometimes be preferable to express them in relative terms (for example, as percentages of the initial employment in the respective region).

The observed changes in Pennsylvania employment in each industrial sector can be split into two components. If durables employment in Pennsylvania had increased at the national rate of 9.61 percent, it would have grown to 787 thousand in 1963 (an increase of 69 thousand). Similarly, Pennsylvania nondurables employment would have increased by 13 thousand if the state’s share of the national total had been maintained. The entire set of results can be summarized as follows for this illustrative case (all figures in thousands):

ENDNOTES

1. Some of the material in this chapter is adapted from E. M. Hoover, ‘Some Old and New Issues in Regional Development" (paper presented at the International Economic Association Conference on Backward Areas in Advanced Countries, Varenna, Italy, August-September 1967). The conference proceedings were published in E. A. G. Robinson (ed.), Backward Areas in Advanced Countries (London: Macmillan, 1969; New York: St. Martin’s Press, 1969).

2. E. M. Hoover, The Location of Economic Activity (New York: McGraw-Hill, 1948), p. 242.

3. The nonwhite population in the United States has been more urban than the white population since some time in the 1950s.

4. The best-known statement of this concern over the implications of slow population growth in America is Alvin Hansen’s presidential address to the American Economic Association in 1938, "Economic Progress and Declining Population Growth," American Economic Review, 29, 1 (1) (March 1939), 1-15. However, Hansen, unlike some of the other contributors to that discussion, did not advocate incentives to higher fertility as a solution.

5. Data from the annual Census sample survey indicate that over the twelve-month period March 1980 to March 1981, about 1 American in 6 moved to a different house, 1 in 16 to a different county, and 1 in 37 to a different state. See U.S. Bureau of the Census, Current Population Reports, Series p-20, No. 377, Geographical Mobility: March 1980 to March 1981 (Washington, D.C.: Government Printing Office, 1983), Table 2, p. 8.

6. See the discussion in Chapter 10 on real income.

7. For a more detailed discussion of regional policy objectives, see Charles L. Leven, "Establishing Goals for Regional Economic Development," Journal of the American institute of Planners, 30, 2 (May 1964), 99-105; Thomas Wilson, Policies for Regional Development, University of Glasgow Social and Economic Studies, Occasional Paper No. 3 (Edinburgh and London: Oliver & Boyd, 1964); Wilbur R. Thompson, A Preface to Urban Economics (Baltimore: Johns Hopkins University Press, 1965), Chapter 5; Benjamin Chinitz, "Appropriate Goals for Regional Policy," Urban Studies, 3, 1 (February 1966), 1-7; and Harry W. Richardson, Regional Economics (Urbana: University of Illinois Press, 1978), Chapter 9.

8. Named after the Italian economist and sociologist Vilfredo Pareto (1848-1923), a pioneer analyst of economic welfare.

9. A challenging and brilliant discussion of the actual and supposed benefits and costs of regional growth is E. J. Mishan, The Costs of Economic Growth (London: Staples Press, 1967). An illustration of the conflicting interests involved in regional prosperity is the report that in June 1969, an association of residents of Hawaii implored convention visitors from the mainland to curb their spending, because such spending raises living costs for the residents. The state of Oregon attracted attention (and perhaps ironically a few additional migrants) in the early 1970s when its governor and other high officials enunciated opposition to in-migration, rapid population growth, and excessive tourism. More and more cities and towns are now seeking constitutional and effective means of limiting growth.

10. Louis Winnick, "Place Prosperity vs. People Prosperity: Welfare Considerations in the Geographic Distribution of Economic Activity," in Essays in Urban Land Economics, in honor of the sixty-fifth birthday of Leo Grebler (Los Angeles: University of California, Real Estate Research Program, 1966). For a much deeper exploration of the place prosperity concept and its relation to regional and national policy objectives, see Marina von N. Whitman, "Place Prosperity and People Prosperity: The Delineation of Optimum Policy Areas," in Mark Perlman, Charles Leven, and Benjamin Chinitz (eds.), Spatial, Regional, and Population Economics (New York: Gordon and Breach, 1972), pp. 359-393.

11. See Matthew Edel, "'People’ versus 'Place’ in Urban Impact Analysis," in Norman J. Glickman (ed.), The Urban Impacts of Federal Policies (Baltimore: Johns Hopkins University Press, 1980), pp. 175-191.

12. Richardson, Regional Economics (Urbana: University of Illinois Press, 1978), p. 146.

13. John H. Cumberland, A Regional Interindustry Model for Analysis of Development Objectives," Papers of the Regional Science Association, 17 (1966), p. 93. For a contrary view see James R. Rinehart and William E. Laird, "Community Inducements to Industry and the Zero-Sum Game," Scottish Journal of Political Economy, 19, 1, (February 1972), 73-89.

14. See Pittsburgh Regional Planning Association, Economic Study of the Pittsburgh Region, 3 vols. (Pittsburgh: University of Pittsburgh Press, 1963). The first volume, Region in Transition, traces the economic history of the region and diagnoses its position as of the time of writing. The second volume, Portrait of a Region, by Ira S. Lowry, focuses on population trends and the geographic patterns of activities within the region. The third volume, Region with a Future, assesses prospects for Further change and suggests some appropriate directions for policy. A concise summary of the study’s approach and findings is E. M. Hoover, "Pittsburgh Takes Stock of Itself," Pennsylvania Business Survey, 5, 1 (January 1964), 4-9.

15. See the illustration of shift-share analysis in Appendix 12-1.

16. This question has concerned many researchers. Some of the most notable recent contributions to this literature include William Alonso, "The Economics of City Size," Papers and Proceedings of the Regional Science Association, 26 (1971), 67-83; Harry W. Richardson, The Economics of Urban Size (Boston: D. C. Heath, 1973); and A. M. J. Yezer and R. S. Goldfarb, "An Indirect Test of Efficient City Size, Journal of Urban Economics, 5, (January 1978), 46-65.

17. Werner Z. Hirsch, Urban Economic Analysis (New York; McGraw-Hill, 1973), Chapters 11-12. See also Nibs M. Hansen, Rural Poverty and the Urban Crisis (Bloomington: Indiana University Press, 1970), Chapter 10; and G. M. Neutze, Economic Policy and the Size of Cities (Canberra: Australian National University, 1965).

18. John L. Gardner, "City Size and Municipal Service Costs," in George S. Tolley, Philip E. Graves, and John L. Gardner (eds.), Urban Growth Policy in a Market Economy (New York: Academic Press, 1979), pp. 51-61.

19. This result lends support to arguments made by Tolley to the effect that economies in the provision of public services are related primarily to the density of population in the service area. See George S. Tolley, "Comparing the Gains and Costs of City Growth," in Tolley, Graves, and Gardner, Urban Growth Policy in a Market Economy, pp. 25-34.

20. See George S. Tolley, "The Welfare Economics of City Bigness," Journal of Urban Economics, 1, 3 (July 1974), 324-345, where the relationship between prices and compensation for urban disamenities is discussed in terms of a "wage multiplier." See also, Oded Israeli, "Externalities and Intercity Wage and Price Differentials,’’ in Tolley, Graves, and Gardner, Urban Growth Policy in a Market Economy, pp. 159-194.

21. See Tolley, "Welfare Economics of City Bigness," pp. 335-338.

22. The external diseconomies argument would also have more weight if it could be established that personal and business migration from large cities is somehow more inhibited than the contrary flow, or that state and national governments are consistently subsidizing locations in large cities at the expense of taxpayers elsewhere.

23. See Richardson, Economics of Urban Size, p. 129; and J. Vernon Henderson, "Effect of Taxation of Externalities on City Size," in Tolley, Graves, and Gardner, Urban Growth Policy in a Market Economy, pp. 91-97.

24. Glenn E. McLaughlin, "Industrial Diversification in American Cities," Quarterly Journal of Economics, 45 (November 1930), 131-449.

25. A cogent discussion of the effects of specialization and business unit size on regional economic resilience is Benjamin Chinitz, "Contrasts in Agglomeration: New York and Pittsburgh," American Economic Review, 51, 2 (May 1961), 279-289.

26. For details on this story, see Pittsburgh Regional Planning Association, Region in Transition, vol. 1 of the Economic Study of the Pittsburgh Region (Pittsburgh: University of Pittsburgh Press, 1963).

27. For a thorough analysis of the spatial distribution of federal outlays (including defense expenditures) over the period 1970-1976, see Georges Vernez, "Overview of the Spatial Dimensions of the Federal Budget," in Norman J. Glickman (ed.), The Urban Impacts of Federal Policies (Baltimore: Johns Hopkins University Press, 1980), pp. 67-102.

28. The People Left Behind, Report of the President’s National Advisory Commission on Rural Poverty (Washington, D.C.: Government Printing Office, 1967), p. xi (italics added). Professor Mary Jean Bowman justly observed that the inclusion of the italicized words could make this recommendation "a prescription for national disaster." "Poverty in an Affluent Society," in Neil W. Chamberlain (ed.), Contemporary Economic Issues (Homewood, Ill.: Irwin, 1969), p. 99.

29. Notably John B. Lansing and Eva Mueller, The Geographic Mobility of Labor (Ann Arbor: Survey Research Center, University of Michigan, 1967).

30. Ibid., p. 77. See also the reference to the "Beale hypothesis" on p. 284 above.

31. In April 1969, the U.S. Supreme Court ruled that a state may not impose residency provisions "for the purpose of inhibiting migration by needy persons into the state." Up to that time, most states had denied welfare assistance to applicants with less than a year’s residence. There are still large differentials in the level of such benefits, however, which substantially affect interregional migration.

32. See, for example, Niles Hansen, "Policies for Nonmetropolitan Areas," Growth and Change, 11, 2 (April 1980), 8.

33. Relevant also is the increasing nonlinearity of costs of transfer with respect to distance— resulting from the relatively greater speed of long-distance transport and communication and the increasing importance of time as an element in costs of transfer for goods, people, services, and information. With respect to communication, personal travel, and shipment of an increasing range of goods, the added time required for an additional several hundred miles is often less than the time required for the first 10 miles.

34. David L. Brown, "Spatial Aspects of Post-1970 Work Force Migration in the United States," Growth and Change, 12, 1 (January 1981), 9.

35. Equal opportunity here basically refers to the removal of employment discrimination based on color, sex, or any other personal characteristics not relevant to work performance. Open entry refers to the removal of inappropriate restrictions on occupational or geographic mobility imposed by union rules and employment agreements regarding hiring, apprenticeship, union membership, or transfer of seniority and pension rights.

36. For an overview of policies concerning interregional labor mobility in France, Great Britain, and the Netherlands, see Norbert Vanhove and Leo H. Klassen, Regional Policy: A European Approach (Montclair, N.J.: Allenheld, Osmun, 1980), pp. 371-379; and P. B. Beaumont, "An Examination of Assisted Labour Mobility Policy," in Duncan Maclennan and John B. Parr (eds.), Regional Policy (Oxford: Martin Robertson, 1979), pp. 65-80.

37. The theory and policy of growth centers was developed in Europe and particularly in France, in the 1950s, before attaining much currency in the United States. The broader term "growth poles" does not always have a spatial meaning and is quite loosely used. It refers sometimes to larger developed regions that include centers and sometimes to specific industry complexes, activities, or even single large installations that play a strategic role in sparking new development. Two titles from the large literature on this subject are: J. R. Boudeville, Problems of Regional Planning (Edinburgh: University Press, 1966); and Niles M. Hansen, French Regional Planning (Bloomington: Indiana University Press, 1968). For Further references and discussion of growth-center concepts and development strategy, see Gordon C. Cameron, "Growth Areas, Growth Centres and Regional Conversion," Scottish Journal of Political Economy, 17, 1 (February 1970, 19-38; and Richardson, Regional Economics (Urbana: University of Illinois Press, 1978), Chapter 7.

38. New York Times, 8 November 1969. Some of the same considerations have been involved in programs in Pittsburgh and other cities to eliminate isolated pockets of settlement on steep hillsides where costs of maintaining streets and other public services are inordinately high. The properties are acquired by the city and all structures removed, with appropriate landscaping and planting.

39. See Brian J. L. Berry, Growth Centers in the American Urban System, vols. 1 and 2 (Cambridge, Mass.: Ballinger, 1973); and John B. Parr, "Growth Poles, Regional Development, and Central Place Theory," Papers and Proceedings of the Regional Science Association, 31 (1973), 173-212.

40. See Harry W. Richardson, "Growth Centers, Rural Development and National Urban Policy: A Defense," International Regional Science Review, 3, 2 (Winter 1978), 133-152.

41. Such limited empirical evidence as is available concerning the spread effects associated with interindustry linkages suggests that they are very weak. See Niles M. Hansen, "An Evaluation of Growth-Center Theory and Practice," Environment and Planning A, 7, 7 (November 1975), 827.

42. See Thompson, Preface to Urban Economics; and Neutze, Economic Policy.

43. See Upper Great Lakes Regional Commission, Growth Centers and Their Potentials in the Upper Great Lakes Region (Washington, D.C.: Upper Great Lakes Regional Commission, May 1969), prepared by Brian J. L. Berry. Fox’s ideas are set forth in "Agricultural Policy in an Urban Society," American Journal of Agricultural Economics, 50, 5 (December 1968), 1135-1148, and in "A New Strategy for Urban and Rural America," Appalachia, 2, 10 (August 1969), 10-16. Fox’s proposal is to use functional economic areas on the order of a quarter of a million population—not only as units into which to aggregate small communities and rural territory for planning, development, and administrative purposes, hut also as units into which large metropolitan areas might be subdivided. In short, he argues that "our metropolitan areas are too large and our rural communities too small for effective government, a creative social life, and an efficient community."

44. Brian J. L. Berry et al., Metropolitan Area Definition: A Re-evaluation of Concept and Statistical Practice, U.S. Bureau of the Census, Working Paper No. 28 (Washington, D.C.: Government Printing Office, June 1968).

45. See Hansen, "Evaluation of Growth-Center Theory," pp. 826-827.

46. See Harry W. Richardson, "Growth Pole Spillovers: The Dynamics of Backwash and Spread," Regional Studies, 10, 1 (1976), 1-19.

47. The Appalachian Regional Commission (see section 12.7.1) made a sample survey of the gross migration of workers (using Social Security records) between 1960 and 1964, and found that "the migrant workers displayed a distinct tendency to move relatively short distances. Most migrants from the center of Appalachia migrated to nearby areas, still within Appalachia. Migrants from the fringes of Appalachia were more likely to move to the ring of territory surrounding the Region, or to other parts of the United States... All outmigrants increased their income substantially, especially those leaving the central part of Appalachia, where hardcore unemployment has been most severe." Appalachia, 2, 8 (May 1969), 14-15.

The executive director of the Appalachian Regional Commission observed in 1967 that there are large colonies of recent migrants from Appalachia in Chicago, Cleveland, Cincinnati, and a number of other cities outside the region, while "there are no Appalachian neighborhoods in Pittsburgh (despite the fact that it is the only major metropolis in the area." Some explanation for this surprising fact, as he pointed out, lies in the structure and relatively low growth rate of the Pittsburgh economy. But this example is useful in suggesting that centers well outside the boundaries of a depressed or backward area can provide employment opportunities to people living in such an area. Ralph R. Widner, "Experiment in Appalachia," Pittsburgh Business Review, 37, 3 (March 1967), 1-15.

48. William H. Miernyk, "An Evaluation: The Tools of Regional Policy," Growth and Change, 11, 2 (April 1980), 3.

49. U.S. Department of Commerce, 1.981 Annual Report of the Economic Development Administration (Washington, D.C.: Government Printing Office, undated), Foreword.

50. As a condition of eligibility, both redevelopment areas and economic development districts are required to prepare approved plans for their development.

51. A redevelopment area can be eligible with as few as 1500 people (or 1000 in the case of an Indian reservation).

52. Except in Appalachia, where the commission itself exercised the functions performed by the EDA for the other economic development regions.

53. What is said here applies to New England as a whole and to the three Southern states of that region, where four-fifths of its people live. Much of northern New England, by contrast, does share many of the economic characteristics of backward low-income rural areas in other parts of the country.

54. John Friedmann, "Poor Regions and Poor Nations: Perspectives on the Problem of Appalachia," Southern Economic Journal, 32, 4 (April 1966), 472. Similar sentiments are expressed in Niles M. Hansen. Rural Poverty and the Urban Crisis (Bloomington: Indiana University Press, 1970).

55. This approach was apparently first used by Daniel B. Creamer in 1942 in U.S. National Resources Planning Board, Industrial Location and National Resources (Washington, D.C.: Government Printing Office, 1943), and was expounded and used on a large scale by Edgar S. Dunn, Jr., in Harvey S. Perloff, E. S. Dunn, Jr., E. E. Lampard, and R. F. Muth, Regions, Resources, and Economic Growth (Baltimore: Johns Hopkins University Press, 1960). The basic exposition of the method is in that source. What we are calling the "mix component" corresponds to Dunn’s "net proportionality shift," and our "competitive component" corresponds to his "net differential shift." The method was later applied still more extensively in the U.S. Department of Commerce by Dunn and others, and comprehensive tabulations have been prepared and published for U.S. Census regions and smaller areas. See Lowell D. Ashby, Growth Patterns in Employment by County, 1940-50 and 1950-60 (Washington, D.C.: Government Printing Office, 1965).

The logic and usefulness of the shift-share approach were attacked by David B. Houston, "The Shift and Share Analysis of Regional Growth: A Critique," Southern Economic Journal, 33, 4 (April 1967), 577-581. A rejoinder in defense was given by Lowell D. Ashby in the same journal: "The Shift and Share Analysis: A Reply," 34, 3 (January 1968), 423-425. A subsequent critique is H. James Brown, "Shift and Share Projections of Regional Economic Growth: An Empirical Test," Journal of Regional Science, 9, 1 (April 1969), 1-18. For a thorough review of this literature see Benjamin H. Stevens and Craig L. Moore, "A Critical Review of the Literature on Shift-Share as a Forecasting Technique," Journal of Regional Science, 20, 4 (November 1980), 419-437.

56. Industry Group 19 (ordnance and accessories) has been omitted from our computations, since data for that industry group for Pennsylvania were not available. Minor adjustments in the totals have been made to take account of rounding-off errors in adding up the figures for individual industry groups to arrive at the subtotals for durable and nondurable goods industries. Basic data were taken from the 1963 U.S. Census of Manufactures reports.