Printer Friendly

Patterns of growth in metropolitan and nonmetropolitan areas; an update.

Patterns of Growth in Metropolitan and Nonmetropolitan Areas: An Update

IN the 1980's thus far, the growth of personal income and population in metropolitan (metro) areas exceeded that in nonmetropolitan (nonmetro) areas. The 1979-83 average annual rate of growth in personal income in metro areas was 9.16 percent, compared with 7.93 percent in nonmetro areas, and the average annual rate of growth in population was 1.08 percent, compared with 0.88 percent (table 1).

A SURVEY article 2 years ago, using estimates through 1981, signaled a shift in metro-nonmetro growth patterns between the 1970's, when nonmetro growth exceeded metro growth, and the 1980's.1 On the basis of estimates through 1983, the differentials between the metro-nonmetro growth rates are larger than those reported in the earlier article. The earlier article also reviewed the forces at work in the shift that had occurred in the growth patterns between the 1960's and 1970's. Each of these decades had been widely depicted in exaggerated terms--the 1960's as the decade of the "exploding metropolis,' the 1970's as the decade of the "emptying metropolis.' The article noted that the sharp shift in growth patterns in the 1970's was largely a population migration phenomenon; underlying employment and earnings patterns were more continuous. Moreover, the population reversal was confined to the four highly urbanized regions--New England, Mideast, Great Lakes, and Far West.

1. Daniel H. Garnick, "Shifting Patterns in the Growth of Metropolitan and Nonmetropolitan Areas,' SURVEY OF CURRENT BUSINESS 63 (May 1983): 39-44.

The earlier article, in testing hypotheses associated with the exaggerated depictions, found that the forces at work in each decade could not be expected to dominate indefinitely. Rather, underlying self-limiting forces appeared to be at work, restraining the continuation of extremes in area growth and decline as envisioned in the hypotheses and thus making extrapolations of decade-long trends unreliable. The article also suggested, however, that the increasing importance to the national economy of international trade and investment might cancel some of the effect of self-limiting forces in reducing differential area growth rates.

This article updates the review of metro-nonmetro patterns for the 1980's thus far; contrasts the patterns of the 1980's with the earlier patterns, focusing on national industrial developments and the industrial composition of areas; presents subsequent research on the increasing variability of regional area growth rates during the 1970's and the early 1980's; and addresses the possibility of cyclical reinforcement and possible exaggeration of shifts in longer term growth patterns.

Patterns of area growth

Table 2 shows, in addition to the average annual rates of growth in total personal income and population, the average annual rates of growth in earnings and in earnings excluding farm and manufacturing. (Earnings are the sum of wages and salaries, other labor income, and proprietors' income). Growth rates are shown for the United States and regions, by nonmetro and metro areas (the sums of nonmetro countries and of metro counties, respectively) and by size class of the latter, for three time-spans --1959-69, 1969-79, and 1979-83.2 The choice of years for the first two timespans is based on national business cycle peaks, with the aim of separating trend from cyclical changes. The last year is the most recent data available. (See "County and Metropolitan Area Personal Income' in the April 1985 SURVEY.)

2. The estimates for 1959-69 and 1969-79 differ somewhat from those in the earlier article because of revisions of the estimates and changes in the classifications of metropolitan areas. In this article, the Metropolitan Statistical Areas accord with the Office of Management and Budget classification introduced in 1983.

Table 2 shows that, thus far in the 1980's, metro area growth exceeded nonmetro area growth in total personal income, population, and earnings in the Nation as a whole and in all but three highly urbanized regions-- New England, Mideast, and Great Lakes, where population growth in nonmetro areas continued to exceed that in metro areas. This pattern contrasts with that of the 1970's, when nonmetro growth exceeded metro growth in the three measures in the Nation as a whole and in the four highly urbanized regions. The pattern of the 1980's resembles that of the 1960's, when metro growth exceeded nonmetro growth in the three measures in the Nation as a whole and in all but two highly urbanized regions --New England and Great Lakes, where personal income growth in nonmetro areas slightly exceeded that in metro areas.

Industrial bases of area growth.

The last four columns in table 2 show growth rates in total earnings and in earnings when farm, manufacturing, and farm and manufacturing earnings are excluded from total earnings. Comparison of growth rates in total earnings and in earnings excluding an industry shows the combined effect of that industry's growth rate and its relative importance (weight) on the total. Farm earnings are excluded because farming is predominantly a nonmetro industry; manufacturing earnings are excluded because manufacturing has affected metro and nonmetro growth patterns differently. Other industries, for the most part, are consumer-service industries, which respond to--rather than shape--area growth, or industries that tend to be concentrated geographically. Mining and selected recreation- and retirement-related industries in nonmetro areas and selected service industries in metro areas are examples of the latter.

The following discussion of nonmetro and metro growth patterns since the 1960's focuses on, often by reference to these columns of the table, the effects on area growth stemming from agricultural employment decline, manufacturing import competition, sharp price movements, and national business cycles.

Nonmetro areas.--In the 1960's, as had been the case since World War II, nonmetro areas were characterized by continued declines in farm employment. These declines led directly to large-scale population outmigration from nonmetro areas in all regions. As the 1960's progressed, however, the pool of redundant farm workers diminished, and increasing employment in growing nonfarm industries in these areas slowed the net population outmigration.

With continued improvements in their access to national markets, nonmetro areas began to benefit increasingly from the manufacturing industries' dispersion as they sought lower cost locations and diversification insurance against strikes and other potential impedances to access to supplies and markets. In table 2, these developments are seen as farm earnings subtracting almost one-half of a percentage point from the growth rate of total earnings in nonmetro areas nationally and manufacturing earnings adding almost one-half of a percentage point.

By the 1970's, nonmetro areas were characterized by net inmigration of population, reflecting growing job opportunities. Manufacturing earnings and employment continued to grow faster in nonmetro areas than in metro areas. The effect of the declining value of the dollar on international trade mitigated the growing threat from imports to labor-intensive manufacturing jobs that tended to concentrate in nonmetro areas. Other nonfarm, nonmanufacturing industries, however, grew even faster than manufacturing, partly spurred, early in the 1970's, by increased migration to recreation and retirement communities and by boom conditions in fuel and other natural resource industries.

In the 1980's, all the major factors that had contributed to the growth of nonmetro areas had reversed. Farming income was down in 1983 because of drought, but a longer term financial crisis related to high real interest rates, falling crop prices, and declining land values led to an increase in the rate of attrition of farmers. Mining and related industrial activities slumped. Labor-intensive manufacturing jobs declined in the face of increased competition from foreign producers who were advantaged by the sharp strengthening of the dollar.

Metro areas.--Although a number of cities had declining population in the 1960's, the population of their suburbs grew, reflecting continued employment growth in the metro area as a whole. By the 1970's, however, suburban growth was insufficient to offset the accelerating decline of cities, especially in the long-established industrial areas of the New England, Mideast, and Great Lakes regions. An industrial shakeout was in progress: it began in New England and then spread to the Mideast and, more recently, the Great Lakes. The shakeout was particularly felt by the older manufacturing cities, and falling employment led to population outmigration to areas in the South and West, both metro and nonmetro.

In the 1980's thus far, the "big story' about growth--apart from high-technology and defense-related manufactures--has been the rapid growth in services, particularly in what may be called producer services. These business, financial, and professional services tend to concentrate in major metropolitan areas. In addition, the Economic Recovery Tax Act of 1981 tended to encourage capital-intensive development in large metro areas--office building, for example. As described for the four highly industrialized regions, these general factors affecting metro area growth in the 1980's were overlaid by region-specific ones.

In the Great Lakes region, growth rates worsened in the 1980's. Durable goods manufacturing industries were particularly hard hit: first, by the two recessions, and then, during the current expansion, by a disproportionate increase in imports, partly reflecting the strong dollar. Even with an agreement with Japanese auto manufactures to limit shipments to the United States, recovery in the motor vehicle industry did not bring back the former peak manufacturing workforce; some other durable goods industries with less protection against imports fared even worse. With the recovery in employment constrained, metro areas with heavy concentrations in durables manufacturing and related industries were subject to high unemployment and population declines. The earnings growth rate in the Great Lakes region was only onehalf that of the Nation as a whole; the shortfall from the national average was mainly attributable to the large weight of comparatively disadvantaged durable manufactures.

In contrast, growth rates improved in the 1980's in the New England and Mideast regions, except for those metro areas with industrial compositions similar to those of declining metro areas in the Great Lakes. The growth rate of manufacturing earnings lagged behind that of total earnings-- less so in New England than in the Mideast or in comparison with the national average. In New England especially, but also in the Mideast, high-technology and defense-related manufacturing growth rates exceeded the national average. Total earnings growth rates in these regions also exceeded the national average. Although population decline appears to have bottomed out in a number of the older cities, suburban growth has slowed. Improved job opportunities apparently have not taken up all entrants to the workforce and the workers made redundant by the earlier industrial shakeout in most of the metro areas, and net population outmigration has continued.

In the Far West, metro areas benefited from increased defense expenditures and a consequent turnaround in aerospace industries from their relatively depressed levels of the 1970's, as well as the general growth in microelectronics production.

Summary.--As noted earlier, the growth industries thus far in the 1980's, apart from high-technology and defense-related industries, have been the service industries, particularly producer services, which concentrate in major metro areas. Conversely, the laggard and decreasing industries, apart from durable manufactures, were farms, coal mining, oil and gas extraction, metal mining, and other natural resource-related industries-- all of which are concentrated in nonmetro areas and had contributed significantly to the higher growth rates of those areas in the 1970's. These industrial patterns account for much of the shift in favor of metro areas in the 1980's.

Growth rate variability

The shift of metro-nonmetro growth patterns does not imply that the 1980's constitute a return to the patterns of the 1960's. Unlike the 1960's, the 1970's and the early 1980's both show substantial regional variability in the rates of growth in earnings and population (table 3). The standard deviation of the average annual rates of regional growth in earnings was 0.517 in the 1960's; in the 1970's, it increased to 1.738, and in 1979-83, it increased further to 2.116. After excluding the farm and manufacturing industries, which are volatile, the standard deviation still shows a large jump from the 1960's to the 1970's, but a slight fallback in the 1980's: from 0.519, to 1.703, and then to 1.619. Population growth rates also show increased regional variability. In the 1960's, the standard deviation was 0.517, and it jumped to 1.017 in the 1970's and to 1.153 in the 1979-83.(3)

3. However, provisional State population estimates for July 1, 1984, indicate a return to much less regional variability in population growth rates. From mid-1983 to mid-1984, the standard deviation for regional rates fell to 0.458--less than that for the average annual rates in the 1960's. One factor in the reduced regional variability is that population growth turned positive in each of the Great Lakes States--notably in Michigan and Ohio, both of which are highly specialized in motor vehicle production.

The jump in regional growth rate variability between the 1960's and 1970's largely occurred as the southern regions and Rocky Mountain region went from moderately-above-average and moderately-below-average rates of growth in population and earnings, respectively, to well-above-average rates, and as the northeastern and Great Lakes regions went from moderately-below-average to well-below-average growth rates. Regional variability remained high or increased in 1979-83 as the pickup in earnings growth rates in the northeast regions was offset by further declines in the Great Lakes and Plains regions relative to those in the southern and Rocky Mountain regions.

Metro and nonmetro area growth rate variability also jumped after the 1960's. The standard deviation of rates of area growth in earnings more than doubled from the 1960's to the 1970's, and increased about one-half again in nonmetro areas and by one-sixth in metro areas in the early 1980's. Excluding farm and manufacturing earnings results in, in the 1970's, even more variability for nonmetro areas and less variability for metro areas and, in the 1980's, less variability for both metro and nonmetro areas. The difference in metro-nonmetro area growth rate variability, after excluding farm and manufacturing earnings, reflects the larger weight of mining industries, which are volatile, in nonmetro areas and of services industries, which are less volatile, in metro areas. Standard deviations of rates of area growth in population have been much higher in metro areas than in nonmetro areas in the 1960's and 1970's, but have tended to converge in the 1980's as area growth rate variability increased more in nonmetro areas than in metro areas.

The sharp increase in variability of regional and area growth rates in the 1970's and early 1980's reflects national and international economic developments to a large degree. Large fluctuations in relative prices of goods and increased international competition for manufactures took place in both timespans. The shocks associated with the former particularly affected nonmetro growth patterns, but--as will be discussed in the next section-- in opposite directions in the two time-spans. The increased international competition affecting nondefense manufacturing industries appears to be a continuation of developments already exhibited in the 1960's, but with a significant difference that is related to the geography of industrial dispersion.

In the earlier article, it was noted that the geography of industrial dispersion was increasingly becoming worldwide. One aspect was that industries that had earlier dispersed, mainly to the nonmetro areas of the United States, in search of diversification insurance and lower cost production were dispersing to even lower cost foreign areas. This development reflected in part the diminishing competitive advantages for the labor-intensive manufactures in which the nonmetro areas had become specialized. Textiles and other nondurables manufactures are a case in point. Production once located in metro areas in the Northeast was displaced by relocation of manufacturers, mainly in nonmetro areas in the Southeast; this production subsequently experienced some displacement by Asian production. Declining lumber and wood manufacture in the nonmetro areas of the Far West and Rocky Mountain regions reflected other diminishing competitive advantages. First, the Southeast has been displacing the western regions in most domestic markets in the Nation because access to desirable timber stands is easier in the Southeast. Second, in part because of the strong dollar, the western regions have lost to Canada their competitive advantage in supplying neighboring midwestern markets and Pacific-rim nations.

Cyclical effects

Although the initial and terminal points of the timespans discussed in this article correspond with business cycle peaks (except 1983), business cycles may have reinforced or even exaggerated shifts in trends between the timespans as well as in increasing regional growth rate variability. One kind of reinforcement may be from the sharp commodity price movements in the 1970's and 1980's thus far. Sharp price movements and their attendant effects on swings in the marginal efficiency of capital probably reinforced the shifts in area growth patterns between these time-spans.

The effect of commodity price movements began in the early 1970's, with sudden, sharp price increases for fuels, farm products, and other raw materials--all concentrated in nonmetro areas. These price increases stimulated investment in these and related activities. Boom conditions in the favorably endowed areas contrasted with the slump in metro areas that was concentrated in smokestack industries in 1974-75. Starting with the industrial slump in the mid-1970's, however, raw materials prices began to fall. In the 1980's, after adjustment for general inflation, prices of raw materials virtually collapsed, except during the surges associated with the second oil price shock. Supplying and other directly and indirectly related industries were stimulated and then depressed by the effects of the sharp price movements on the raw materials industries. Farm machinery, chemical fertilizers, and regional banks as well as farms were particularly affected by crop price movements; mining and oil field equipment and services, banks, as well as the primary-producing industries were affected by fuel price movements. Moreover, construction and other consumer and producer services were affected. The general boom and bust conditions in the affected nonmetro areas is likely to have accelerated and possibly exaggerated the shifts in the longer term area patterns.

The current boom in office building construction in major metro areas is also likely to be succeeded by collapse. This course is suggested by the prevailing very high vacancy rates and the very large additions to floor space that are coming on stream. However, the speculative building cycle will not necessarily affect metro areas in the same way that the collapse of raw materials prices affected nonmetro areas. In nonmetro areas, high raw materials prices affected nonmetro booms, and the collapse in prices engendered generalized crises. In most metro areas, the office building boom did not tend to result in generalized booms. The outcome of collapsing marginal efficiency of investment in office buildings (declining floor space rental rates), apart from decreased construction, is likely to be increased growth of service industries in metro areas relative to nonmetro areas, because of the increased availability of relatively cheap floor space in the metro areas--an important budget item in many service industries. Further, high vacancy rates in metro areas may slow the movement --noted in the earlier article--of "backroom' service operations to nonmetro areas and, perhaps, preserve for a few years the growth advantages of metro areas in most aspects of the growing producer services industries. These developments appear likely because, given the durability of the capital in place in the metro areas and the relatively large weight of the fixed costs associated with floor space rental, variable cost advantages-- such as lower wage rates for office workers--in nonmetro areas may be offset for some time.

Table: 1.--Total Personal Income and Population: Average Annual Growth Rates, 1959-69, 1969-79, and 1979-83

Table: 2.--Total Personal Income. Population, and Earnings: Average Annual Growth Rates, 1959-69, 1969-79, and 1979-83

Table: 3.--Standard Deviations of Average Annual Growth Rates of Population and Earnings, 1959-69, 1969-79, and 1979-83
COPYRIGHT 1985 U.S. Government Printing Office
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1985 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Garnick, Daniel H.
Publication:Survey of Current Business
Date:May 1, 1985
Previous Article:Sources of change in the federal government deficit, 1970-86.
Next Article:Regional and state projections of income, employment, and population to the year 2000.

Related Articles
County and metropolitan area personal income, 1980-82.
County and metropolitan area personal income, 1982-84.
Growth in metropolitan and nonmetropolitan areas: an update.
Accounting for regional differences in per capital personal income growth: an update and extension.
Variations among local economic regions.
Comprehensive revision of local area personal income estimates, 1969-90.
Local area personal income.
Town and country. (Labor Month in Review).
Brookings study reveals new view of sprawl. (Special Report).
Nonmetropolitan area occupational data now available.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters