The Bureau of Business Research The University of Texas atAustin May-June 1979 U.S. Oil Geography in 1990: Scenarios and Implications 65 Uses and Misuses of Metropolitan Area Employment and Unemployment Numbers 70 Hispanic Fertility in Texas 79 The City as Magnet: Residential Choice in Dallas 81 The Socioeconomic Status of Texas Cities and Suburbs 85 The Federal Economic Stimulus Package: Effect on Large Texas Cities 89 Manufacturing in Texas: How Recession Proof? 95 The Industrial Screening Matrix: A Tool for Industrial Developers 98 Longview-Marshall: An Industrial Center 103 Tyler: Promising a Rose Garden 106 Texas BUSINESS The Bureau of Business Research • ev1ew The University of Texas at A us tin Vol. 53, No. 3, May-June 1979 U.S. Oil Geography in 1990: Scenarios and Implications 65 James W. McKie and E. Victor Neimeyer Uses and Misuses of Metropolitan Area Employment and Unemployment Numbers 70 Lorna Aldrich Monti Hispanic Fertility in Texas 79 Alejandro Velez The City as Magnet: Residential Choice in Dallas 81 Donald A. Hicks The Socioeconomic Status of Texas Cities and Suburbs 85 Alfred J. Watkins The Federal Economic Stimulus Package: Effect on Large Texas Cities 89 Joseph E. Pluta Manufacturing in Texas: How Recession Proof? 95 Lowell Dworin The Industrial Screening Matrix: A Tool for Industrial Developers 98 Deborah G. Jones, Douglas J. Manifold, and Jean S. Spencer Longview-Marshall: An Industrial Center 103 Charles P. Zlatkovich Tyler: Promising a Rose Garden 106 Charles P. Zlatkovich Cover design: Eje Wray Bureau of Business Research Charles C. Holt, Director Review Staff Lorna A. Monti, Editor Lois Glenn, Publications Manager Charles F. Dameron, Jr., Managing Editor Mary Jo Powell, Editorial Assistant John A. Burghardt, Data Consultant Mildred Anderson and Jean Hall, Data Compilation Daniel P. Rosas, Printing Coordinator Joan Farnham, Compositor Subscription rate: $9 .00 per year. Single copy: $2.00. Pub­lished six times a year. Second-class postage paid at Austin, Texas. Publication number 540-400. ISSN 0040-4209. Copyright © 1979, Board of Regents, University of Texas System. Address manuscripts and communications to Texas Busi­ness Review, Bureau of Business Research, P.O. Box 7459, Austin, Texas 78712. Telephone: 512-471-1616. Texas Business Review is indexed in Marketing Information Guide and Public Affairs Information Service and is avail­able on microfilm from University Microfilms. U.S. Oil Geography in 1990 Analysis of alternative futures for the oil industry reveals that costs of transportation, reflecting the locations of markets, sources, and existing refineries, tend to pull addi­tions to U.S. refining capacity away from the Texas Gulf Coast toward more eastern points on the Gulf and toward inland sites. But economies arising in the Texas complex of refineries and petrochemical plants-with associated tech­nical skills, support activities, and dependent industries­offset this disadvantage. The development of a Texas superport would also add to the attraction of the Texas Gulf. These conclusions arise from analyzing alterna­tive geographic patterns of the oil industry in 1990, pat­terns that began shifting markedly during the 1970s. The Pattern before 1970 After World War II the geographical flow of domestic crude oil and refined products in the United States took a form that changed little until the 1970s. Oil produced west of the Rockies was refined and consumed there; im­ported crude oil made up the growing difference between domestic supply and demand in the area after 1960. Oil produced in the midcontinental area, including West Texas, flowed north to midwestern refineries; the remainder of the oil needed in that region was largely supplied by imported crude oil from Canada and by refined-product shipments from other regions. Oil produced in the Gulf Coast region of Texas and Louisiana moved to Gulf ports, where it was either refined or transshipped to crude-oil carriers for movements to East Coast refineries. Although some refined products from the Gulf Coast were consumed within the region, most were shipped to other regions by pipeline, James W. McKie is Professor of Economics, University of Texas at Auatin. E. Victor Niemeyer is Project Manager with the Systems Program, Electric Power Research Institute. The authors thank the Center for Energy Studies for supporting the study and James Story for preparing the data for the maps. For a more detailed version of thil atudy, see E. Victor Niemeyer and James W. McKie, U.S. Oil Geography in 1990: Scenarios and Implications for Economic Policy, Policy Study no. 5 (Austin: Center for Energy Studies, Uni­venlty of Texas at Austin, October 1978). Scenarios and Implications James W. McKie E. Victor Niemeyer barge, and tanker; the principal destination was the East Coast. The growing deficit on the East Coast was met by im­ports of crude oil (mostly from Venezuela until after 1970) and of refined products from the Caribbean and Europe. Changing Influences after 1970 During the 1970s, the following major developments have influenced oil geography in the United States: • Production of crude oil in the lower forty-eight states has declined while demand has continued to grow, resulting in a rapidly growing reliance on imported oil, with the Middle East as a major supplier. Crude oil has ceased to move from the U.S. Gulf Coast to the East Coast in any significant quantities. The East Coast has become almost wholly dependent on im­ported crude oil, and foreign oil has begun to move into Gulf Coast ports, which are now import points rather than export points for interregional move­ments. Some refined products still move from the Gulf Coast to the East Coast by pipeline. • The Canadian government has decided to phase out crude-oil exports to the United States (primarily to the Midwest) and to divert supply into eastern Cana­da, displacing imports into Canada. • Alaskan oil has begun to flow from Prudhoe Bay, potentially eliminating oil imports into the western region of the United States and making some addi­tional oil available east of the Rockies. • The environmental protection movement has limited the development of refining capacity on the East Coast, thereby promoting the flow of already-refined products into that region from abroad and from the U.S. Gulf Coast. The Pattern in 1976 A detailed view of the geography of the U.S. petroleum industry in 1976 is presented in figure 1, showing imports and interdistrict flows of crude oil and refined products. Figure 1 Movements of Refined Products, 1976 (Thousand barrels per day) 194 ,. Crude 011 ref ned) In 1976 the districts east of the Rockies were closely linked by the flow of crude oil from the Gulf and the Rocky Mountains to the Midwest and by the flow of re­fined products from the Gulf to the East Coast and the Midwest. Because there was little movement of either crude oil or refined products between the West Coast and the rest of the country, the markets there were relatively indepen­dent. These markets are expected to become integrated with the rest of the country in the 1980s. Scenarios for 1990 In projecting scenarios for 1990, we are concerned with minimizing the cost of meeting predicted demands for crude oil and refined products in the country. Four factors must be considered: the cost of transporting refined prod­ucts and crude oil between selected pairs of locations, the the cost of refining crude oil into finished products at dif­ferent locations, the cost of importing crude oil and fin­ished products at the different locations, and the costs and limits on production of crude oil. The crucial numerical assumptions in this analysis in­volve the costs of refining crude oil into products, the cost of imported crude oil and refined products, the supply of crude oil, and the 1990 demand for refined products for I : East Coast II: Midwest 111: Gulf Coast and West Texas IV: Rocky Mountains V: West Coast different locations. Changes in these numerical assumptions alter the geography of the U.S. petroleum industry, and alternative scenarios can be compared accordingly. In the 1990 base case, demand for refined products is assumed to be 23.6 million barrels per day (bpd), a moder­ate estimate. Domestic production of crude oil and natural gas liquids is assumed to be 12.7 million bpd. Other major assumptions in the base case are as follows: • Imports of petroleum products are not limited. • Two Gulf Coast superports with unlimited capacities are located off the Texas and Louisiana coasts. These ports enjoy a cost advantage of 50 cents a barrel over transshipment through the Bahamas. • Total Alaskan production is 3 million bpd. • Additional refining capacity is allowed on the East Coast. There, as well as in all the other districts, incre­ments to capacity are limited to 300 thousand bpd or an amount equal to the capacity existing in 1977, whichever is greater. This constraint, which prevents installation of large refineries in regions that have not been significant refining centers, is based on the assumption that such regions would not tolerate politically an increase in capacity beyond one large refinery. • A crude-oil pipeline with a capacity of 500 thousand bpd links southern California and West Texas. This Figure 2 Movements of Refined Products, Base Case, 1990 (Thousand barrels per day) pipeline corresponds to the one that has been pro­posed by Sohio. • Another pipeline links northern Washington State with the Midwest. Two maps (figure 2) outlining the base case show several notable changes from the 197 6 case (presented in figure 1 ). The principal changes are as follows: • 1,400 thousand bpd of crude oil and refined products flow from the West Coast to the Midwest. • Southern and northern transcontinental routes are used to move the surplus of crude oil on the West Coast to the East. • Imports to the West Coast cease. • The flow of imports into the Gulf Coast increases great­ly (though much of it is on its way to the Midwest). • Both superports are operated at approximately their initial design capacity. Together they account for 77 percent of crude-oil imports and 56 percent of total oil imports. • Interdistrict flows of refined products remain at ap­proximately their 1976 levels. • Increased demand on the East Coast is met by slightly higher shipments from the Gulf Coast, a cessation of the net flow from the East Coast to the Midwest, in­creased use of existing capacity on the East Coast, and, finally, an additional 1,200 thousand bpd of im- Movements of Crude Oil, Base Case, 1990 (Thousand barrels per day) I : East Coast II: Midwest 111: Gulf Coast and West Texas IV: Rocky Mountains V: West Coast ports. From 1976 to 1990 the percentage of refined products in total imports increases slightly from 24 to 27 percent. • No refineries are added on the East Coast (though the use rate is assumed to increase), and all new refineries are built inland. • The increase in imported refined products is not very large when compared to demand. From 197 6 to 1990 imported refined products increase from 10 to 12 percent of total demand. The assumed lack of increase in refining capacity on the East Coast (when refining is not constrained) and the com­paratively small increase in imported refined products need some explanation. In Factors Affecting U.S. Petroleum Refining, published in 1973 , the National Petroleum Coun­cil reported that the cost of meeting demand on the East Coast was lowest through importing products from offshore (Caribbean) refineries, somewhat higher through refining imported crude oil on the East Coast, and highest through refining imported crude oil on the Gulf Coast and then shipping the products to the East Coast. The advantage of the offshore refineries is principally a result of lower taxes and fewer environmental constraints. While the costs of re­fining are lower along the Gulf Coast than on the East Coast, the savings in refining costs do not outweigh the added transportation costs. Only variable costs need be recovered for the continuing operation of existing domestic refining and pipeline capa­city along the Gulf Coast. This analysis does not indicate that any additional pipeline or refining capacity should be added to this system; importing refined products is the cheapest means of meeting demands beyond current capa­cities. The implication is that the competition from unre­stricted product imports to the East Coast would keep some pipelines and refineries from fully recovering their fixed costs. New domestic refining capacity will be added inland be­cause products from these locations do not compete direct­ly with refined products that are shipped from the coastal refineries on the Gulf Coast through pipelines linking the East and Gulf Coast. In scenarios with limitations on im­porting refined products, incremental refining capacity is located on the East Coast and inland but not on the Gulf Coast, a result significant to Texas and Louisiana. A Texas Superport The question of where superports should be built is not likely to be settled by reference to costs of shipment and locus of demand for imported petroleum. Political decisions appear already to have ruled out superports on the East Coast. Serious intentions to build them are confined at present to the Gulf Coast. Construction of one superport off the Louisiana coast, LOOP, has already begun. A second, originally called Sea­dock, is planned off the Texas coast. The initial plans for the Texas port, however, were aborted when its promoters­oil-importing companies-balked at the environmental and economic regulations that the U.S. government proposed for it. During the summer of 1978, the Texas legislature authorized a state-owned superport (the Texas Deep­water Port Authority, or TDPA) financed by state-issued bonds to be amortized with income from port operations. The decision on whether to build the Texas deepwater port could have significant effects on the patterns of trans­portation and refining if they respond to relatively small differences in costs, as we assume they do. In our analysis that includes a Texas superport, we pro­ject movements in 1990 of imported crude oil through the Texas superport at 3,647 thousand bpd, while 2,295 thou­sand bpd are projected to move through the Louisiana superport. These two ports are considered to account for all the crude-oil imports into the United States except for the 1,795 thousand bpd transshipped in the Bahamas and imported onto the East Coast. (It is cheaper to unload crude oil by large barges, called lighters, on the East Coast or to transship it from the Bahamas than to unload it at a Gulf Coast superport and transship it by any means to the East Coast.) The indicated amounts are within the design capacities of the two superports. What happens if the Texas superport is not built? We consider two possible cases. The first case assumes that the Louisiana superport is restricted to a capacity of 3 ,000 thousand bpd in 1990 and that no other superport will be constructed. About 220 thousand bpd of surplus West Coast crude oil shifts from the northern to the southern route to replace some crude oil that is moved to feed the 3,780 thousand bpd of exist­ing refining capacity on the Texas Gulf Coast, which is assumed to continue. The rest of the crude-oil requirements for the Texas Gulf Coast is supplied partly by imports through the Louisiana superport and partly by crude oil brought in by small tankers from Mexico or transshipped through the Bahamas. Total imports of crude oil through the Gulf Coast are less than in the base case by about 1,900 thousand bpd. The higher prices paid for crude oil on the Gulf Coast decrease the ability of refiners there to compete in East Coast markets, a situation that could deflect their output to the Midwest instead. The deficiency is made up by the increased importing of crude oil and refined products directly to the East Coast. Crude-oil imports through the East Coast increase by 1,030 thousand bpd, but almost all of it moves on to the eastern Midwest. The importing of refined products to the East Coast increases by 910 thou­sand bpd. A third of these extra imported refined products are passed on to the eastern Midwest, while the rest make up for reduced shipments of refined products through the pipelines into the East Coast from the Gulf Coast. On balance, loss of the Texas superport results in imports going to the East Coast instead of to the Gulf. The second case assumes that the capacity of the Loui­siana superport is unlimited and that its lower costs are passed along in lower prices for crude oil. An equivalent assumption is that several competitive superports are built along the Louisiana-Mississippi-Alabama Gulf shore. The Texas superport is again deleted. In this case, all imports arriving in supertankers are unloaded in Louisiana and vicin­ity. We cannot predict how much Mexican oil would be un­loaded from smaller tankers in Texas, but the remainder of what is required by Texas refineries would have to be trans­shipped from the Louisiana superports or other points, such as the Bahamas. Whether the oil comes from Louisiana, other transshipment points, or directly from Mexico, the landed cost of incremental oil supply in Texas would be higher than on the eastern Gulf Coast. The Eastward Pull Shifts resulting from the deletion of the Texas superport in either case reflect a fundamental fact that must influence any policy regarding petroleum facilities on the Texas Gulf Coast in the near future: the geographical pull for imports is toward the east. Based on an analysis completely uncon­strained by existing refinery facilities and pipelines but pre­serving the Gulf Coast as the only locus for superports, the best location for a single superport with unlimited capacity would be in the vicinity of Mobile and Pensacola. If pipe­lines and refineries could be relocated without cost, that point would be the optimal site, considering transportation costs of the resulting system. The assumptions are artificial, but the results do indicate a moderate disadvantage in trans- Figure 3 Movements of Refined Products without Texas Superport, 1990 (Thousand barrels per day) port costs for Gulf Coast points (for both input and output) that increases in a westerly direction. To a large extent this disadvantage is offset by econo­mies of agglomeration in the Texas Gulf Coast complex of refineries and petrochemical plants, with its strong infra­structure of technical skills and support activities and its superstructure of dependent industries. The basis of the growth of this complex has been the movement of Texas­produced oil and gas toward coastal points. In the future, imports to the Gulf Coast will supply all incremental re­quirements and will increasingly replace Texas crude oil and natural gas production as inputs to this coastal industrial complex. Associated with the general growth of imports is the conversion of future petrochemical plants from natural gas liquids to naphtha input stocks, also largely to be im­ported or refined from imported oil. Hence there will be a tendency for incremental investment in new refineries and other petroleum-oriented industries to be located farther east along the Gulf, unless the Texas Gulf region can con­tinue to offer countervailing advantages. Implications for the Texas Gulf If the Gulf Coast is to become dependent on oil imports, and if superports are to be built in Louisiana and perhaps Movements of Crude Oil without Texas Superport, 1990 (Thousand barrels per day) I: East Coast II: Midwest 111: Gulf Coast and West Texas IV: Rocky Mountains V: West Coast elsewhere-both developments that are well underway at the present-it is in the economic interest of Texas to have a superport and to facilitate its early construction by all reasonable means. Furthermore, that interest does not conflict with any larger national interest, since the net resource cost of oil transportation and refining would be at least as low with the Texas superport as without it. That statement does not imply that the environmental costs of the superport should be ignored; they will probably be handled by federal regulations that apply to other super­ports as well. Nor does it imply that the Texas superport would need a permanent subsidy, though it might require initial state financing that can be recovered through user charges. To be without a superport might cause loss of some of the forward momentum of industrial development and shift incremental additions to refinery, petrochemical, and pipeline capacity toward the east. New specific taxes on refinery and petrochemical plants in Texas would increase the vulnerability of the Texas Gulf complex to competitive sites in the new petroleum environ­ment. Such taxes might well be a decisive deterrent to in­cremental investment in the Texas Gulf region. The func­tioning of that complex during the next decade or so may depend strongly on preservation of lower costs and superior advantages from concentration of skills and technology over competing areas that have some advantages oflocation. The Uses and Misuses of Metropolitan Area Employment and Unemployment Numbers Lorna Aldrich Monti The analyst of metropolitan economic trends often turns to the only current and comprehensive measures of such phenomena available-monthly estimates of employment totals and unemployment percentages produced by state employment service agencies, such as the Texas Employ­ment Commission, which is under the technical direction of the U.S. Bureau of Labor Statistics. The best available does not mean perfect, however, and users of these num­bers must be aware of their origins and limits. Analysts using these figures for forecasting, service plan­ning, and market research are not the only ones interested in their accuracy; city, county, and state officials have an urgent concern because over $16 billion in federal funds, most notably for employment and public works programs, were distributed by formulas using these estimates in 1977. 1 Numbers originally designed for economic analysis are used to allocate dollars among state and local governments; this development mushroomed in the 1970s because Congress wrote laws that use specific state and local statistics to help allocate money. What do these statistics show about Texas SMSAs? Five­year trends in the estimates of nonagricultural employment and percentage of the labor force unemployed for Texas SMSAs are displayed in the accompanying graphs. Valuable basic knowledge about the relative position of SMSAs is immediately apparent: for example, before the 1973 oil embargo and subsequent recession, total employment in the Houston SMSA was below the Dallas-Fort Worth total by approximately 13,000 persons, but by the end of 1978 the two were nearly equal. The 197 4-197 5 recession caused a sharp dip in the Dallas-Fort Worth employment graph, while it remained practically invisible on the Houston graph. These basic trends can be distinguished even though the estimates are subject to larger relative errors than the corresponding national reports of employment totals and unemployment percentages. The level of accuracy of the statistics on metropolitan areas is most critical in comparisons of unemployment rates Loma Aldrich Monti is Editor, Texas Business Review. in different SMSAs. For example, the unemployment situation indicated by Corpus Christi's 5.3 percent rate in November 1978 does not necessarily differ from the one indicated by Galveston-Texas City's 6.0 percent rate. On the other hand, Laredo's 13.5 percent rate undoubtedly indicated a situation different from that shown by Austin's 3.3 percent rate. Because they must be based on administrative records that were created for purposes other than estimating em­ployment and unemployment, the metropolitan area esti­mates can be used for general comparisons and trends but not for precise comparisons. The national estimates, on the other hand, are produced from a monthly survey of over 60,000 households called the Current Population Survey, which is designed to develop labor market information. The two methods intertwine because the national definitions of employment and unemployment are used in the procedure for producing administrative records and because employ­ment and unemployment totals for states and large metro­politan areas taken from the national survey are used to revise the administrative-record data. Annual numbers for large SMSAs and states are derived from the national survey and are used to adjust the level of monthly estimates that come from the administrative-record procedure. Levels of Accuracy The national survey levels of accuracy are such that a change of 0.2 percentage points is required before a change in the national unemployment rate reflecting more than inescapable "noise" (random variation) from the survey process can be detected. For example, an analyst cannot conclude that a reported movement from 6.0 percent to 6.1 percent means that the actual national unemployment situation changed. Applying the same standard to the sur­vey results for states and metropolitan areas would imply that only a change of 1 percentage point or more-from, say, 6.0 percent to 7 .0 percent-could be taken as more than statistical noise.2 As discouraging as this information might be to an ana­lyst or recipient of funds allocated by unemployment statistics, the amount of noise in the monthly SMSA unem­ployment rates (which are numbers derived not from the national survey but from the administrative-record estima­tion procedure) is even greater. Because the administrative­record procedure was not designed according to statistical methods whose accuracy can be gauged, errors in the method cannot be estimated, although hints can be derived from comparisons with the decennial census and the survey results. The average difference between the administrative­record estimates and the survey for Texas between 1970 and 1976 amounted to -18.4 percent of survey unemploy­ment,3 or an underestimate approaching one-fifth of the total. Since 1973 the estimates have been revised to con­form to the survey results; so the level of Texas unemploy­ment is not currently underestimated. Because survey re­sults are available only for Houston, Dallas-Fort Worth, and the state as a whole, however, there may still be large errors in the estimates for other Texas SMSAs, although the administrative-record estimate of the state total is revised to match the state survey total. The employment and unemployment graphs suggest that the estimates do contain noise. Some fluctuations that re­ peat in an SMSA every year could easily be the result of seasonal patterns not caught in the estimating process, but the rough graphs also indicate that random effects can appear. National Survey Concepts Each month, during the week containing the twelfth day of the month, the national Current Population Survey queries over 60,000 households about the employment status of members 16 years old and over. The survey de­ fines two types of employed persons: those who were paid for working, even if for only one hour, during the reference week; and those who were temporarily absent from their jobs or businesses. Paid work is not an absolute require­ ment ; unpaid work in a family business for fifteen hours or more constitutes employment.4 The definition of unemployment also sets minimum standards. Any job search activity in the four weeks before the survey, such as asking friends about availability of jobs or registering with an employment service, defines a person as unemployed provided he or she was also available for work during the survey week. Looking for work is not an absolute requirement; persons laid off from a job and waiting to be recalled or expecting to report to a new job within thirty days are also defined as unemployed. 5 Individuals who do not meet the minimum standards of either definition are not classified as members of the labor force. The definition of the civilian labor force also ex­ cludes inmates of institutions and members of the armed forces. The very broad standards for employment and unem­ ployment that are used in the survey do not correspond to popular conceptions of employment and unemployment (perhaps fifteen hours of paid work and ten or more hours spent seeking work), but they do cover a range of economic activity that would escape stricter definitions. Because both employment and unemployment definitions are minimal, is is difficult to know if the official unemployment rate for the civilian labor force is greater or smaller than the one that would result from standards closer to popular concep­tions. Further controversies about the definitions center on people who do not look for work because they think it is unavailable, although they want work and are available. Administrative-Record Approach Despite the large number of households interviewed for the Current Population Survey, the number of surveyed households in each jurisdiction falls short of requirements for statistical accuracy . One estimate puts at 200 million An administrative-record approach is used for estimating monthly unemployment rates in Texas metropolitan areas. annually the number of interviews required to produce acceptably reliable survey results each month for the 6,000 geographic subdivisions for which unemployment numbers are required by laws. Conducting this many inter­views would cost $4 to $5 billion.6 Such numbers are im­practical, and the amount of interviewing required would lead to resentment of large-scale government inquiries into personal details. More reasonable costs are associated with annual, rather than monthly, surveys, but the total costs still appear impracticable. The cost of statistical surveys dictates the use of an administrative-record approach. The first step in such an approach is to reconcile the definition of unemployment from the Current Population Survey with the requirements of eligibility for unemployment insurance in the state pro­grams. The differences between the definitions used in various states contribute to the inaccuracies of estimates from the administrative-record approach. Use of survey results as benchmarks corrects the levels of unemploy­ment for states, such as Texas, where definitional problems would otherwise lead to underestimation or overestimation. This administrative-record method must reconcile unem­ployment definitions and then estimate separately: (1) un­employed persons last working in industries presumed covered by unemployment insurance, (2) unemployed per­sons in noncovered industries, and (3) unemployed persons who have been out of the labor force and have begun to look for work.7 Unemployed persons last employed in non­covered industries and new entrants to the labor force must be estimated using various relations, such as past experience of entrants in the labor force compared to past experience of other unemployed persons, employment levels in non­covered industries, and national unemployment rates. Some of the relations are as old as the last decennial census, and all introduce some error into the process. The years further from the census will contain larger errors. use of past relations to make current estimates, and the record of month-to-month inaccuracies in carrying forward annual benchmarks for SMSA employment estimates. A few simple reminders, however, enable users to benefit from the timeliness and frequency of the data without making serious errors of judgment. Since the figures are adjusted for levels annually, annual changes and trends are more reliable than monthly trends. Month-to-month changes should be viewed cautiously until The estimates are better for tracking trends in one SMSA than for making comparisons between SMSAs. Because the errors are consistent over time, changes in unemployment rates are more accurate than the levels. 8 For example, if two SMSAs had unemployment levels of 4.0 and 5.0 percent, the difference might be entirely a sta­tistical artifact. If, however, one SMSA's rate moved over several months from 4.0 to 5.0, it would be likely that the basic unemployment situation had changed. The estimates are better for tracking trends in one SMSA than for making comparisons between them. The labor market problem of unemployment receives much more public attention and analysis than labor market successes in employment. Forecasting, market research, and public service planning all benefit from estimates of total employment as a measure of the level, trend, and composi­ tion of economic activity. The administrative-record method generates such data monthly for most SMSAs in Texas. Employment totals, like unemployment totals, must be estimated by using the unemployment insurance programs as the base and other information to fill in gaps. Results from the national survey for the state and for large SMSAs are used to correct levels of employment once a year. Annual estimates of all employment are made once a year from reports on employment and wages in the unem­ployment insurance program, supplemented by indepen­dent estimates for government, nonprofit organizations, and similar entities. These levels are carried forward month to month through a partial reporting system, which in­cludes establishments that employ 41 percent of the workers covered by the estimates. The largest month-to­month errors are in the mining, contract construction, and service industries. 9 Implications for Users Analysts who are using estimates of employment and un­employment for forecasting, planning, or market research should be mindful of the variations among the surveys, the differences in definitions between the Current Population Survey and state unemployment insurance programs, the a trend has been established over several months. Differ­ences in unemployment rates among SMSAs must be sev­eral percentage points before they can be interpreted as in­dicators of basic differences in labor market conditions among SMSAs. Monthly changes in employment in the con­tract construction, mining, and service industries should be interpreted with caution. The user who keeps these reserva­tions in mind can still examine the composition of employ­ment across industries for several SMSAs, note the trends over time in the level of employment and unemployment in one SMSA, and discover changes in the composition of em­ployment within one SMSA over time. These uses of the data will enable the analyst to chart trends in employment growth, look for trends in the com­position of employment that may signal changes in the growth trend of employment, and discover past suscepti­bility of an SMSA to conditions elsewhere in the nation or in key industries. The estimates are an imperfect tool that, if used correctly, can produce solid analysis. Notes 1. Janet L. Norwood, "Reshaping a Statistical Program to Meet Leg­islative Priorities," Monthly Labor Review, November 1977, p. 6. 2. Harold Goldstein, State and Local Labor Force Statistics, Na­tional Commission on Employment and Unemployment Statistics Background Paper no. 1 (Washington, D.C.: Government Printing Office, May 1978), pp. 17-18. 3. Ibid., p. 58. 4. Bureau of Labor Statistics, U.S. Department of Labor, Handbook of Labor Statistics 1977, Bulletin 1966 (Washington, D.C.: Govern­ment Printing Office, 1977), pp. 1-2. 5. Ibid. 6. Goldstein, p. 64. 7. Martin Ziegler, "Linking CPS Unemployment Estimates With Administrative Records on Unemployment," in U.S. Department of Commerce and Bureau of the Census, Interelationships Among Estimates, Surveys and Forecasts Produced by Federal Agencies, Small Area Statistics Papers Series GE-41, no. 4 (Washington, D.C.: Government Printing Office, June 1978), p. 37. 8. Ibid., p. 38. 9. Goldstein, p. 39. Employment Trends in Texas SMSAs Two kinds of labor data are graphed for each SMSA. Nonagricultural employment (in thousands of employees) is indicated by a solid line and measured on the left-hand scale. Unemployment (percentage of the labor force) is indicated by a broken line and measured on the right-hand scale. The graphs are grouped according to SMSA size. The range of nonagricultural data may differ from chart to chart. DALLAS-FORT ~ORTH 1300-r----------..,..-22 --NONAGR EHPLOYHENT N -----------· PERCENTAGE UNEHPLOYED 20 0 N A 18 p ~ 1200 E 16 R I c c E u NL 14 T T A u G 12 E~ 1100 L u 10 N E E H Hp 8 p L L 01000 ,', . 0 y I I I I\ I\ I '-'\ 1' 1\1 :1 .. 6 y 1 H ''J ''! 'J ' ,, E I ',' 1 11 I \ I l ,, E •' "• ', ' ,.. ' '' \ ,1 \ I \ I ' ·', DN I\ f\._I : I \I \I\ I , 1 \ \ I 1 I ti \ 4 1 T I,,.I V ' 11 v' 900----~~-~-~--2 1974 1975 1976 1977 1978 MAY-JUNE 1979 HOUSTON 1300-.-------------..-22 --NONAGR EHPLOYHENT N 20 -----------· PERCENTAGE UNEMPLOYED 0 N A 18 p G1200 E R I 16 R cc E u NL 14 T T A u G R1100 12 E A L u 10 N E E H H p 8 p L L 01000 0y 6 y H ~ ., I "\ I\ E E ~ I 1 I I 1, I • I I I I I '" I\ ', DN I I\ •I I t I I f \ JI I ' 4 ,._,,,•, ,,,,,,,..., '.. ',• \•'' 1',, T : 11:: .,.1 II # I 1: I II I .' , ~ 900---~~-~-~--+-2 1974 1975 1976 1977 1978 73 Employment Trends in Texas SMSAs SAN ANTONIO AUSTIN N360-.------- -------~22 N220--.--------------~22 0 0 ---NONAGR EMPLOYMENT N 20 p N 29 p ------······ PERCENTAGE UNEMPLOYED A A E E G G 16 R 18 R R R c c I I 16 E c c 16 E N N U340 U200 T T L L 1'4 A 14 A T T G G u u 12 E 12 E R R A A u u L L 10 N 10 N E E E320 E180 8 H 8 HH H p pp p L LL L 6 0 6 0 0 y 0 .... " y y y ' t\ ~,', .... : "''\ ,• ', ....... ,.. 4 E H •," ••• 4 E I ,, • \ • '•' \ : \ .... , ,, H \ {-'·,~~-·· ,. ,. \ ,'"'\. • D DE E \ 'v' : 'W N300-+---.---,----~---.----+-2 N160-r-'·~·-.---.-----,.---r---+-2 T T 1974 1975 1976 1977 1978 1974 1975 1976 1977 1978 EL PASO BEAUMONT-PORT ARTHUR-ORANGE N180--.--------------.....-22 Nl80--.--------------~22 0 0 ---NONAGR EMPLOYMENT ---NONAGR EMPLOYMENT N N 20 p 20 p ------------PERCENTAGE UNEMPLOYED ------------PERCENTAGE UNEMPLOYEDA A E E G G 18 R 18 R R R c c I I 16 E c c 16 E N N U160 Ul60 T T L L 1'4 A 14 A T T G G u u 12 E R 12 E A R A u u L L 10 N 10 N E E E140 E140 8 M 8 HM p H pp p L LL L 6 0 6 0 0 y 0 y y y 4 E 4 E H H D D E E N120-;---.---.-----,---.---+-2 N120-+----r-L--.----~---.----r-2 T T 1974 1975 1976 1977 1978 1974 1975 1976 1977 1978 74 TEXAS BUSINESS REVIEW Employment Trends in Texas SMSAs LUBBOCK CORPUS CHRISTI N129-.....-------------"""T"""22 N 90-.--------------~22 0 0 ---NONAGR EMPLOYMENT--~NONAGR EHPLOYHENT N N 20 p20 p -----------· PERCENTAGE UNEMPLOYED-----------· PERCENTAGE UNEMPLOYED A A EE G G 18 R18 R R R cc I I 16 E 16 E c c NN u 80 U110 T T L L 14 A 14 A T T G G u u 12 E 12 E R R A A u u L L 10 N 10 N E E E 70 E109 8 H 8 H M p H p p p L L L L 6 0 6 0 0 0 y yy y 4 E 4 E H H D D E E N90~--~--...-----""""T---...----+-2 N 60-+---~--...----..---.---+-2 T T 1974 1975 1976 1977 1978 1974 1975 1976 1977 1978 GALVESTON-TEXAS CITY AH ARILLO N80--.--------------..-22 N 80.....,...-------------~22 0 0 ---NONAGR EMPLOYMENT ---NONAGR EMPLOYMENT N 20 p N 20 p -----------· PERCENTAGE UNEMPLOYED -----------· PERCENTAGE UNEMPLOYEDA A E E G G 18 R 18 R R R c c I I 16 E c 16 E c N N u 70 u 70 T T L L 14 A 14 A T T u 12 G E u 12 G E R R A A u u L L 10 N 10 N E E E 60 8 M 8 H H p pp L L L I ,,.... ,\,..' 'I ,, . ', 1\ 6 0 6 0 I, ' : ', ~ ,-... , • ' •J .. ,./ 0 I I 1" 1 I I,..,. f y y y •a f ~ t I I I II ', u 4 E 4 E H '• . D D E N 50-+----.----...----..---.---+-2 T 1974 1975 1976 1977 1978 1974 1975 1976 1977 1978 MAY-JUNE 1979 75 Employment Trends in Texas SMSAs N HC ALLEN-PHARR-EDINBURG 0 66--.----------~~-----~22 N WACO o 1a~------------------,-22 P ~ ----NONAGR EHPLOYHENT 29 E G -·······-·--PERCENTAGE UNEMPLOYED R R 18 C I E c 18 ~ U 86 A L 14 G T E u 12 R A L 1874 1876 191e 1877 1878 N LONGVIEW-MARSHALL ~ 6Q-..------------------r-22 p A ----NONAGR EMPLOYMENT 29 E G -----------· PERCENTAGE UNEMPLOYED R R 18 C I E c 16 ~ U 56 A L 14 G ~ E 12 R U A 10 N L E 59 E 8 H H ~ p I -.. p \ 1'.. 8 L L .......:' ......,..! •, :'..'".....' 0 0 ~ 4 y y E H 45-4----~---r-----.---...----+-2 D E N 1874 1976 1978 1977 T K:ILLEEN-TEl1PLE 1974 1976 191e 1977 1978 N A G R I 80 c u­ L T ~ 55 A L E H 50 P L 0 y p ----NONAGR EMPLOYMENT 29 E ------------PERCENTAGE UNEMPLOYED R 18 c E 115 N T . A ... 14 G f ,,, .. E 12 ~ / U : : 10 N •: E 8 H P 8 L 0 4 yE H 45-+---~---.---~.----.....----+-2 D E N 1974 1975 1978 1977 1978 T N WICHITA FALLS o 66~----------------.-22 P N A ----NONAGR EMPLOYMENT 29 E G --····-----· PERCENTAGE l.IEWLOYEO R R 18 C I E c 18 ~ U 50 A L 14 G T E u 12 R U A 19 N L E "45 E 8 H H p p 8 L t :'\ .. ,"... /.,,,,,._........... :'.·\_ ,-.... ~ ~ y \ :·\"....: • .... .. -· ,_, ....,r..........._,.\,,_,-·-..: E H "49-+-·---~---.-----..----~---t-2 D E N 1874 1876 1878 1877 1878 T Employment Trends in Texas SMSAs N ODESSA ~ se.........-------------------.-22 P A ----NONAGR EMPLOYMENT E 29 G ------------PERCENTAGE UNEMPLOYED R R 18 C I E C N U '45 I 6 T L A T 1'4 G U E R 12 A U L 10 N ~ E E 8 H H P P !\ 6 L L • • 0 0 :\.. ,·"v: \.,'\ ,'\ :•,,"" •'', •' 4 Y y ..........• ,... '-,J '........,..../ .. , E H 3S--+----,.----T""----.-------.---t-2 D E N 197" 1975 1976 1977 1978 T N TYLER TEXARKANA ~ s0-.----------------.......--zz P A ----NONAGR EMPLOYMENT ze E G ------------PERCENTAGE UNEMPLOYED R R 18 C I E c 16 ~ U '4S A L 1'4 G ~ E R 12 A U L 10 N ~ E E ~ 8 M H ~ :\ / \,\ 1\ • P p : ~ ,/ •, __ ,: " '"\.., :'"".. ,.., 8 ~ L I \" ""' .. ' " .. :'..' ....,,.·\, ,;-"'\...... 0 ..~ ,.. .._. 'J .... 4 y y \_: E Hss----~--~---~---.------z D E N U~75 1976 1977 1978 197" 1975 1976 1977 1978 T N MIDLAND N SHERMAN-DENISON ~ ..a-...----------------.......--zz P 0 '40-r-----------------,-22 p A ----NONA.GR EMPLOYMENT 29 E NA ----NONAGR EHPLDYMENT E 20 G -----------· PERCENTAGE UNEMPLOYED R G -----------·PERCENTAGE UNEMPLOYED R R 18 C R 18 C I E I E ~ 16 ~ c 16 ~ 35 U 35 A L A T l'4 G L 1'4 G T E u A U R ~~ •\ ,..., A ~ ,, 1 2 u L 19 N A :l\:• 'IV .. •,:'". 19 N ~ 38 12 E L • \' , • \ E E 30 : './ • E 8 H E : '. 8 H P1 25 p • p p ~ 8 L M \/-'·'.... /-.._\ ...... 8 L L :~ ~ ,"..., ~ ~ 0 p '.J '.... 0 0 .... : • ....\.'•:~ .. ....., : ' ... ,.., ' f' ,• '"\ 4 y L •'' 0 4 y y ".' '• I y E H 28-4----~--~---~--~----2 D M 25-+----,.----...----.-----,.------1t-2 D \ : I -; t .. ,' --·',, ,• .._. •...... / ........: " ......• ·-.,,: ""'" E E E N 197'4 1975 U~76 1977 1978 197" 1976 1976 1977 1978 N T T MAY-JUNE 1979 77 Employment Trends in Texas SMSAs BRYAN-COLLEGE STATION N SAN ANGELO 0 35--..~~~~~~~~~~~~~~~-,-22 p ~ ----NONAGR EMPLOYMENT 20 E G ---·-------· PERCENTAGE UNEMPLOYED R R 18 C I E N C 16 T U 30 A L 14 G T E u 12 R U A 10 N L E 25 E 8 M p M .~ 8 L p :\ : ~ ,,j\ , : 0 Lo ,I, ,' \ , ~ : \, ... ,, : " \ /", '"' .. I,,\ I. "'T v A 1 y . ,,,, 4 .. • -,/ ', ,,~ ,, ......~,·· E ' ..,,/ • ..... ........ ...... ......, H 20~~~~~~~...--~~-.-~~-r~~----1--2 D E 197'4 1975 1978 1977 1978 1974 1975 1978 1977 1978 N T N LAREDO ~ 35.....,.~~~~~~~~~~~~~~~~~-22 p A ----NONAGR EMPLOYMENT E 20 G ------·----· PERCENTAGE UNEMPLOYED R R 18 C I • E cu l ,.......~ • :\,4, .~.. 18 ~ 30 .~ , • 4 : • '• L 1 • ' I ' / : A T f\ J\ 1,f \,/\ / \ I • I : I 4 G UR ~ ,~ } ~ ' : V.. ~./ ~:'. /' ~ : 12 E ~ : ~ ,' ~ ' . ./ 'l .. , '...,......'\} A 'I\.,; U L "•.; 10 N ~ E E 8 M M p p 8 L L 0 0 4 y y E H 20__,.~~--,,----~~...-~~-.-~~--..~~--1~2 D E 197'4 1975 1978 1977 1978 N T Estimates of employment and unemployment rates are produced by the Texas Employment Commission under the technical direction of the U.S. Bureau of Labor Statistics. These estimates, previously published monthly in the Review, and the graphs that accompany this article are now available in a monthly statistical summary. Readers wishing to re­ceive the summary should direct requests to the Bureau of Business Research. Hispanic Fertility in Texas Hispanic persons lead the population growth in Texas, which itself is in the country's fastest-growing region. Pres­ently totaling at least 18.4 percent of the state's popula­tion, Hispanics have been especially important in the growth of two of the six fastest-growing SMSAs in the state-San Antonio and El Paso-where at least one out of every two persons is Hispanic. Regionally, Hispanics are particularly numerous in the southern and western portions of the state, making up at least half the population in most Texas counties located on or near the Mexican border. Estimates of future Hispanic population growth in Texas depend upon both immigration and natural increase. While it is commonly recognized that public policy may influence the former, fertility is often regarded as a variable that is determined by factors outside the realm of public policy. This notion concerning fertility needs to be reevaluated. Factors That Influence Fertility Rates Hispanic women have had higher fertility rates than women of other ethnic origins have had. In an earlier study Alejandro Velez is Assistant Professor of Economics and Finance, University of Texas at San Antonio. Alejandro Velez of the fertility of U.S. residents of Hispanic origin, I found that the education of the husband, the type of income earned, and the degree of population density help to ex­plain rates of fertility among Hispanics. Using a microeco­nomic framework, I demonstrated that household income and the time available to raise children are crucial factors in a family's decision on how many children to have. It is thus possible to correlate the number of children in a family with the family's resources, education, general knowledge, residential location, and even age. Hispanic Population Growth in Texas Percentage Group 1960 1970 change Hispanic 1,417,811 • 1,840,862 •• Texas total 9,579,677 11,195,415 16.9 *This figure was based on the term Hispanic as denoting persons with Spanish surnames. ••A more comprehensive term, not available in 1960, was used here. Hispanics, however, are still undercounted in the census be­cause of the wording of the questionnaire and because of the illegal alien status of some Hispanic persons. So urces: Texas A&M University, A Decade of Population Change in Texas, 1963; and U.S. Department of Commerce, U.S. Census ofPopulation: 1970-Supp/ementary Report. The educations of the husband and the wife reduce fer­tility in two complementary ways. More education im­proves the effectiveness of contraception and makes indi­viduals more valuable on the labor market-a fact that tends to limit the time they are willing to spend at home raising children. Because raising children is more expensive in urban areas then in rural areas, urban families tend to have smaller fami­lies. Rural families that own farms, however, also tend to have smaller families than other rural residents. Apparently farm owners have more education, higher child-rearing costs, or higher incomes than other rural Hispanics. Predictably, transfer payments increase fertility because they give parents with less education (and a lower cost of raising children) a positive incentive to have more children. Government transfer payments, which increase with the number of children born, are an important factor in in­creased fertility among Hispanics. Declining Fertility Rates Economic forces are at work to slow Hispanic fertility rates in Texas and in the nation. It is encouraging that, ac­cording to the U.S. Bureau of the Census, younger Hispanic married women plan to have roughly the same number of children over their lifetimes as Anglo or Black married women. Public policy aimed at higher educational attain­ment among Hispanics can be considered a powerful influ­ence in reducing fertility rates in Texas. Similarly, while urbanization brings many problems, it reduces family size among Hispanics and should be intelligently encouraged. For those remaining in the countryside, large families and a perennial cycle of poverty need not prevail. Public policy aimed at increasing agricultural proprietorship among His­panics might reduce fertility rates and offset some of the Hispanic Population as Percentage of Total Population, 1970 0 o. 4.9 0 5.0 -9.9 Gl] 10.0. 19.9 lillil 20.0. 49.9 • 50.0 and over Source: Charles P. Zlatkovich, Rita J. Wright, and Robert S. Moore, Texas Focf Book !Austin: Bureau of Business Research, University of Texas at Austin, 1978J, p. 20. factors behind the migration to the big cities in Texas. Finally, welfare programs (as they are presently con­structed) encourage larger Hispanic families, whose children are likely themselves to become charges of the state. Note Many of these conclusions evolve from my earlier study, "An Economic Analysis of the Fertility of Spanish Origin Households in the United States in 1970" (Ph.D. diss., University of Florida, 1977). Average Number of Expected Lifetime Births Per Wife in the United States Age at d ate of survey Race or origin 18 to 39 18 to 24 25 to 29 30 to 34 35 t o 39 Year years years years years years Spanish origin• 1977 2.80 2.26 2 .50 3 .05 3.54 1976 1971 2.85•• 2 .35 •• 2 .4 5•• 3 .2 9 •• 3.47• • Black 1977 2.79 2.09 2 .30 2 .99 3.87 1976 2.80 2 .30 2 .5 1 2 .92 3.58 1971 3.31 2.62 3 .11 3 .71 4.22 White 1977 2.39 2. 14 2 .18 2.42 2 .88 1976 2.42 2 .13 2 .1 8 2 .51 2 .9 5 1971 2 .74 2.3 5 2 .58 2 .94 3. 19 All races 1977 2.43 2.14 2 .20 2.47 2 .9 5 1976 2.45 2 .14 2.2 0 2 .54 2 .99 1971 2 .7 8 2.38 2.62 2 .99 3.26 *Persons of Spanish origin may be of any race. ••Not available. Sources: Bureau of the Census, Fertility of American Women: June 1977 (Advance Report), and Current Population Reports, Series P-20, no. 316, December 1977. From 1974 to 1978, mobile Dallas residents were in­creasingly interested in relocating within the city limits and decreasingly interested in moving elsewhere in Texas. This development may partly be explained by the attractiveness of the Dallas economy within the state, the relative costs of suburban and central locations, and national trends that indicate a preference for central cities among city residents. Five distinct migration and mobility streams characterize the portion of the Dallas city population that expects to change residences in the near future. The first stream, made up of those who move within the Dallas city limits, includes relatively short-distance relocations that cross no political boundaries. The remaining four streams are moves to Dallas suburbs (in Dallas County but outside the city), moves to surrounding counties, moves to locations elsewhere in Texas, and moves to locations outside Texas. National Trends In a nationwide study of movers,1 85 percent of last moves were intrametropolitan, 28 percent were within the same neighborhood, and 57 percent were outside the neigh­borhood in which the movers had lived. Only 15 percent of all last moves crossed a metropolitan boundary. Evidence on the relative sizes of the various migration and mobility streams often contradicts conventional wisdom. Most moves are within the central city, and the two most com­mon relocation destinations are the same neighborhood in the central city and a different neighborhood in the same central city. Relocations to a central city from other metro­politan areas, towns, and rural areas account for nearly 6 percent of last moves; another 3 percent involve the forma­tion of new households. Migration from the suburbs to the central city accounts for 2 percent of all last moves nation­ally. Finally, and certainly out of proportion to the amount of attention it has received, less than 10 percent of all re­locations in the country involve moves from a central city to the suburbs of that city. An estimated 60 percent of all relocations are voluntary and are accounted for by a family's changes in household Donald A. Hicks is Assistant Professor of Sociology and Political Economy, University of Texas at Dallas. A more extensive presenta­tion ofthe data in this study is available from the author on request. The City as Magnet Residential Choice in Dallas Donald A. Hicks size and income, which either necessitate or make possible a move to a new residence.2 According to a national study conducted in 1972,3 nonwhite, low-income, and rental households are more likely to relocate within the same neighborhood than white, high-income, and homeowner households are. Previous residential status (renter or homeowner) has long been a reliable predictor of residential choice. Renters are not only much more likely to move than owners are, but also they are more likely to move short distances and to relocate in the central city. While general dissatisfaction with a neighborhood is asso­ciated with both intended and actual moves, some evidence indicates that specific complaints about such environmental factors as crime, violence, and air pollution do not account for moves out of a neighborhood.4 Relocation Expectations of Dallas Residents Survey results indicate that these national trends are mir­rored in patterns of planned and actual relocations in the Dallas area. Since 1974 the city of Dallas has conducted annual city profile surveys to assess the effects of a wide variety of municipal services. 5 Items on actual and planned residential relocations, as well as on desired locations and destinations, have been included in these surveys. According to the city surveys in 1974, 1976, and 1978, relocation expectations strongly favor moves within the city limits. In 1974, an estimated 25 percent of households in the city of Dallas reported plans to relocate in the next year or two. This ratio of one in four stayed nearly con­stant over the next two surveys, but the choice of destina­tion changed markedly between 1974, a recession year, and 1976 and 1978, both years of improvement in the Dallas area economy. In 1974 every third household expecting to move chose a destination within the city of Dallas, while in both of the later surveys this proportion rose to approxi­mately one in two. Relocations elsewhere in Texas declined in popularity from 30 percent of potential movers in 1974 to 18 and 14 percent in 197 6 and 197 8. This shift may be related to the relative improvement of the Dallas economy, which was not insulated from the 1974-1975 recession as the Houston economy was. From 1974 to 1978, relocations outside Texas declined in popularity from 15 percent to 9 percent, while reloca­tions in Dallas County (outside the city limits) and in sur­rounding counties improved slightly-an average of about one out of five households chose these areas. The reasons given by households for planning or desiring to move were generally consistent across the three surveys. In addition to the desire for larger, newer, or better hous­ing, the desire to be a homeowner was mentioned most fre­quently. Preference for another neighborhood and a desire to move nearer to places of employment were also fre­quently mentioned. One of the least frequently mentioned reasons for moving, however, was schools. Schools were generally viewed as an asset, not a liability, of the neighbor­hood being left behind. The fact that potentially mobile households increasingly turned to residential destinations within the city of Dallas is reflected by vacancy rates in Dallas. In both 1974 and 1976, the vacancy rate for single-family dwellings was steady at 6 percent; between 1976 and 1978 that rate dropped to 4 percent. A sizable part of the demand for such housing among Dallas residents was probably met within Dallas itself rather than in the suburbs. An opposite trend occurred in vacancy rates among multifamily struc­tures, such as apartments and condominiums. Despite very high demand for such housing, especially among young singles and childless couples, supply eclipsed demand by 1978, and many units were empty. The vacancy rate for multifamily structures rose from 8 percent in 1974 and 1976 to 10 percent in 1978. Suburbs may be beginning to lose some of their appeal over central cities. The urbanization of suburbs has brought with it increases in housing costs and local taxes, overbur­dened urban services, and a general flattening of cost-of­living differences between central cities and their suburbs. Thus, added to the general tendency of city residents to prefer to remain in the city is the increasing attractiveness of the central city for those who are not residents. The Dallas metropolitan area has served as an economic magnet in recent years. Actually, the metropolitan periph­ery, not the city of Dallas, has enjoyed most of the benefits and endured most of the burdens accompanying rapid growth. The survey evidence may be a signal that the intra­metropolitan pattern of the benefits and burdens of growth is changing. Trends within Selected Population Subgroups Race, age, income, and past housing experience are sig­nificant determinants of household moving patterns in both national and Dallas surveys. Race or Ethnic Status Black and Hispanic groups mirrored the general pattern of a shift away from relocations outside the Dallas area in 1974 toward moves within the city in 1976 and 1978. In all three years higher proportions of Blacks and Hispanics answering the survey expected to live within the city limits. Thus, the general concentration of minorities in central cities appeared in the Dallas survey together with the more general trend , which Anglos also share, of the growing pop­ularity of the city. Age of Head of Household National evidence indicates that while the age of the head of the household is one of the most useful variables for predicting a move, it is relatively unimportant in pre­dicting the location choice. The increased popularity of the city and decreased popularity of other locations in Texas and beyond Texas are found among all age groups, but with varying intensity. Households with heads younger than age thirty-five are increasingly more interested in central loca­tions than households with heads older than thirty-five are. Total Family Income The basic trends held true for all income groups, al­though the strength of the trend toward central locations was greater for lower-income households. Mobile people in Suburbs may be beginning to lose some of their appeal over central cities. higher-income households ($15 ,000 or more) expected to remain in the Dallas area rather than move to locations else­where in Texas and out of state, but locations outside the city yet within Dallas County were increasingly popular. Tenure Status Both owners and renters in Dallas indicated an increasing tendency to plan to relocate inside Dallas and a decreasing tendency to plan to move outside the Dallas metropolitan area. In any single survey, however, renters will prefer the city of Dallas, while owners will favor destinations in the suburbs, in the larger metropolitan area, or in other parts of Texas. Housing Type Two principal types of households can be defined by the type of housing structure they occupy: those in single­family, detached dwellings and those in multifamily, multi­unit structures such as apartments, condominiums, and townhouses. Households in the multifamily, multiunit structures at all three times chose Dallas as a residential des­tination more often than did households living in single­family dwellings. The pattern is reversed when we examine those who leave the Dallas metropolitan area but stay in Texas. Among both renters and homeowners, interest in relocations within the city of Dallas has increased consider­ably from 1974 to 1978; interest in the suburbs has in­creased slightly; and interest in places outside the metro­politan area and places outside Texas has decreased. Length of Residence in Dallas Generally, those who have arrived in Dallas most re­cently are those most likely to be planning to move. There has been an increase in the proportion of mobile house­holds, regardless of how long they have resided in Dallas, that seek a new location in Dallas. This trend accompanies the declining interest in residential destinations outside the metropolitan area regardless of the length of residence in Dallas. Increasingly, the suburbs have gained in attractive­ness only for those who have been in Dallas fewer than five years or more than ten years. Length of Time at Current Address Preference for places outside the Dallas metropolitan area tends to decrease across time but increase with length of time at current address. One weak, longer-term trend is for surburbs increasingly to lure potentially mobile house­holds. It is apparent, however, that for people expecting to relocate in the near future, other places in Dallas or other places in Texas are more likely destinations than are locations in the suburbs ringing Dallas. Notes 1. Edgar W. Butler, Urban Sociology: A Systematic Approach (New York: Harper and Row, 1976). 2. Janet Abu-Lughod and M.M. Foley, "Consumer Differences," in Nelson Foote et al., Housing Choices and Housing Constraints (New York: McGraw-Hill, 1960), pp. 95-133; and Ronald J. McAllister, Edward J. Kaiser, and Edgar W. Butler, "Residential Mobility of Blacks and Whites: A National Longitudinal Survey," American Journal ofSociology 77 (November 1972): 445-56. 3. McAllister, Kaiser, and Butler, "Residential Mobility." 4. Theodore Droettboom et al., " Urban Violence and Residential Mobility," Journal of the American Institute ofPlanners 37 (Sep­tember 1971): 319-25. 5. Each of the three Dallas Community Profile Surveys that I chose to examine was drawn with standard interval samples based on the 1970 U.S. census block count of dwelling units and updated through records of electrical connections. A two-part questionnaire consisted of a visual exterior housing rating to be made by the interviewer followed by a set of questions for face-to-face inter­views. Although the items on household mobility in the 1974 study were not worded identically to the items in 1976 and 1978, it is unlikely that such slight noncomparability substantially weakens the inference of the longer-term trends. the onlyone: If you buy or sell in Texas: 1978-1979 Directory of Texas Manufacturers Over 14,000 plants listed: name, address, telephone, executive officer, and product descriptions. Volume 1: Volume 2: plants listed by firm name products listed by S.l.C. number and by city and by city within each S.l.C. group. includes complete addresses and an index. $50 plus tax Bureau of Business Research The University of Texas at Austin Box 7459 Austin, Texas 78712 ------,-_____, BBR Publications on Latin America Atlas of Mexico Arbingast, Stanley A., et al. Revised 1975. 164 pp., some maps in color. $20.00. ISBN 87755-187-1. An economic atlas of Mexico. Credit Systems for Small Scale Farmers: Case Histories from Mexico Williams, Simon, and James A. Miller. 1973. 260 pp. $5.00. ISBN 87755-153-7. Studies in Latin American Business No. 14. An examination of those credit operations touching the Mexican campesinos. A Spanish edition is available from Editorial Diana, Mexico. Economic Integration in Latin America: The Progress and Problems of LAFT A Mathis, F. John. 1969. 112 pp. $3.00. ISBN 87755-076-X. Studies in Latin American Business No. 8. The study is con­cerned primarily with the Latin American Free Trade Asso­ciation as the organization expected to determine the suc­cess or failure of the common market in Latin America. A Spanish edition is available from Editorial Diana, Mexico. Industrial Polarization under Economic Inte­gration in Latin America Garbacz, Christopher. 1971. 101 pp. $3.00. ISBN 87755­138-3. Studies in Latin American Business No. 11. The focus is the tendency for an economic union of countries with widely divergent levels of development to result in further extreme concentration of economic activity at a few industrial poles. Industrialization and Employment in Puerto Rico, 1950-1972 Holbik, Karel, and Philip L. Swan. 1975. 82 pp. $3.00. ISBN 87755-208-8. Studies in Latin American Business No. 16. Although de­velopment programs dating from the 1940s have made pos­sible the emergence of Puerto Rico as an industrialized economy, the economic changes have resulted in certain problems, problems that are a major focus in the book. International Tourism and Latin American Development Krause, Walter, and G. Donald Jud. 1973. 74 pp. $3.00. ISBN 87755-176.f:>. Studies in Latin American Business No. 15. The central ques­tions in the work are (1) What is the potential for tourism in Latin America? and (2) What must Latin America do to realize the potential? Mexican Migration and the U.S. Labor Market: A Mounting Issue for the Seventies Briggs, Vernon M., Jr. 1975. 37 pp. $1.50. ISBN 87755­214-2. Studies in Human Resource Development No. 3. An analysis of the effects of U.S. labor, immigration, and border poli­cies on employment and labor problems of the seventies. The Mexican Migration Numbers Game: An Analysis of the Lesko Estimate of Undocu­mented Migration from Mexico to the United States Roberts, Kenneth, Michael E. Conroy, Allan G. King, and Jorge Rizo-Patron. 1978. 33 pp. $4.00. ISBN 87755­228-2. Research Report 1978-1. A detailed account of methods used to enumerate illegal aliens in this country and a critical assessment of this approach. The Mexico-United States Border: Public Poli­cy and Chicano Economic Welfare Briggs, Vernon M., Jr. 1974. 28 pp. $1.50. ISBN 87755­200-2. Studies in Human Resource Development No. 2. A dis­cussion of the effects of U.S . immigration policies. Monetary Accommodation of Regional Inte­gration in Latin America Ziegler, Lawrence F. 1971 . 83 pp . $2.50. ISBN 87755­154-5. Studies in Latin American Business No. 12. The author's focus is the monetary side of the Latin American move­ment toward market integration. Social Class and Consumption Behavior in Sao Paulo, Brazil Cunningham, Isabella C.M., et al. 1976. 177 pp. $4.00. ISBN 87755-258-4. Studies in Marketing No. 23. The authors examined the shopping and consumption behavior of individuals in vari­ous social classes in Sao Paulo. Trade and Industrialization in the Central American Common Market: The First Decade Holbik, Karel, and Philip L. Swan. 1972. 67 pp. $3.00. ISBN 8775 5-167-7. Studies in Latin American Business No. 13 . The authors trace the development of the Central American Common Market and evaluate its results. The United States and Latin America: The Alliance for Progress Program Krause, Walter. 1963. 35 pp. $2.00. ISBN 87755-070-0. Studies in Latin American Business No. 2. A discussion of the political and economic involvement of the United States with Latin America through the Alliance for Progress. Wage Differences between United States and Guatemalan Industrial Firms in Guatemala Maddox, Robert Casey. 1971. 57 pp. $2.00. ISBN 87755­143-X. Studies in Latin American Business No. 10. The wage levels and factors accounting for differences are analyzed. The Socioeconomic Status of Texas Cities and Suburbs Metropolitan areas in Texas differ significantly from metropolitan areas in most other sections of the nation. Not only are they growing more rapidly in population and number of jobs, but there is an even chance that their cen­tral cities rank higher than their surrounding suburbs when measured along several indices of socioeconomic status, such as median years of education and median income. The typical metropolitan area outside Texas contains a relatively poor central city inhabited by people who have lower levels of education, lower incomes, and a greater probability of living below the poverty level than do resi­dents of surrounding suburbs. As a consequence of this intrametropolitan polarization, these suburbs have strenu­ously and repeatedly rebuffed most annexation overtures initiated by their central cities; the incentives for the middle class to flee the central cities in favor of what they perceive to be the relatively safe and problem-free haven in the suburbs have increased; and the central cities, particu­larly in the Northeast, have been left with mounting social problems, greater demands for expensive social services, an eroding tax base, and, in several instances, a last-ditch need to avoid financial insolvency. To date, Texas annexation laws have enabled most of the metropolitan areas in the state to avoid these northeast­ern patterns of intrametropolitan polarization with their attendant deleterious effects on the health and vitality of their central cities. Despite this relatively favorable situa­tion, many of the largest metropolitan areas in Texas have begun to replicate the northeastern pattern as have the Sun­belt SMSAs of New Orleans and Atlanta. Whether or not this situation will become dominant in the 1980s and serve as a harbinger of future urban problems will depend upon the precise array of growth and social policies pursued by central cities in the state. Furthermore, although social disparities between rich and poor will continue to exist, the political strategies formulated to address these inequalities Alhed J. Watkins is Assistant Professor of Govemment, University of Texas at Austin. Alfred J. Watkins will be altered by the fact that both groups will be compe­ting for political advantage within one jurisdiction. Patterns of Intrametropolitan Polarization in Texas Two commonly used measures of socioeconomic status­median school years completed by all persons at least twenty-six years old and median family income-have been employed for comparing the socioeconomic status of thirty central cities in Texas to the status of their suburbs. (Bryan, College Station, and San Angelo have been omitted from this study because the Census Bureau has failed to delineate an urban balance for these metropolitan areas; cities de­fined as metropolitan areas after the release of the 1970 census have also been excluded.) For statistical purposes, the U.S. Census Bureau defines the suburban portion or urban balance of an SMSA as consisting of all communities within a metropolitan area that are outside the central city and satisfy certain minimum standards of density. This defi­nition should distinguish the rural portions of certain met­ropolitan counties from those communities that should properly be regarded as economic and social extensions of the core city. Socioeconomic Status in 1970 Median school years completed and median family in­come highlight the positive aspects of the Texas metropoli­tan areas. When measured by these criteria, fifteen out of the thirty cities included in the study ranked higher than their suburbs in 1970 (see table l ). Included in this list are Abilene, Harlingen, Corpus Christi, Lubbock, Midland, Odessa, Sherman, Denison, Tyler, and Wichita Falls-all relatively small cities. Two university towns, Austin and Waco, have higher values for the education variable than do their suburbs but lower values for median family income. Amarillo and El Paso rank higher than their suburbs in income but lower in education. Significantly, residents in the suburbs surrounding the state's four largest central cities-Houston, Dallas, San Antonio, and Fort Worth-are better educated and have higher incomes than their coun­terparts inside the city limits. For these four cities, as well as for eleven others in the state, the northeastern patterns of intrametropolitan polarization, while not yet at a crisis stage, had emerged nearly ten years ago. While Texas has a fairly even division between SMSAs with higher socioeconomic status in the central cities and those with higher socioeconomic status in the suburbs, the northeastern metropolitan areas have a uniform pattern of deprivation in the central cities and affluence in the suburbs (see table 2). The one exception, Indianapolis, has recently completed an extensive annexation program. Socioeconomic Status in 1960 In comparison to the prevailing situation in the North­east, metropolitan areas in Texas are significantly less po­larized. When the 1970 statistics for Texas are matched with comparable data from the 1960 census, however, the direction of the trends offers little ground for either opti­mism or complacency. Intrametropolitan polarization in­creased dramatically during the 1960s. In 1960, nearly two-thirds of the Texas central cities were more affluent and better educated than their suburbs. By 1970, only 50 per­cent of these central cities exceeded the median income and median education levels of their suburbs. The significant question is whether the 1980 census will reveal a continua­tion of this trend. While there does not appear to be a no­ticeable tendency for the socioeconomic disparities to have widened or attained crisis dimensions, during the 1960s thirteen Texas cities-including Dallas, El Paso, Austin, Waco, Brownsville, Laredo, and Texarkana-dropped below their suburbs in either median education, median family in­come, or both. The Role of Annexation Policies From the perspective of the 1970 census, many SMSAs in Texas are relatively unique entities. Unlike almost every other metropolitan area in the nation, they cannot be char­acterized by the presence of a poor, socioeconomically dis­advantaged central city surrounded by high-status, econom­ically booming suburbs. In large measure, the relatively lenient annexation provisions enacted by the Texas legisla­ture are responsible for enabling the state's central cities Table 1 Socioeconomic Status of Texas Cities and Suburbs, 1960 and 1970 1960 1970 Median school years completed Median family income (dollars) Median school years completed Median family income (dollars) Central cities Central city Urban balance Central city Urban balance Central city Urban balance Central city Urban balance Abilene Amarillo 12.1 12.1 11.9 12.1 5,460 5,877 5,063 5,820 12 .3 12.3 10.3 13.6 7,827 8,867 6,475 8,565 Austin Beaumont Port Arthur Orange Brownsville Harlingen San Benito Corpus Christi Dallas El Paso Fort Worth Galveston Texas City Houston Laredo Lubbock McAJlen Pharr Edinburg Midland Odessa San Antonio Sherman Denison Texarkana Tyler Waco Wichita Falls 12 .0 · 11.3}9.7 11.2 7 .6 } 9.3 6.2 10.7 11.6 11.6 11.2 9.0}11.3 11.4 6 .1 12.0 12.4 11.8 9.5 10.8 12 .1 10.9 11.8 11.9 10.9 7 .6 10.2 12 .0 10.6 11.6 10.6 11.5 6.2 11.8 • 12.4 11.6 9.9 9.9 11.1 10.5 11.6 5,119 5,577}5,659 5,510 3 ,021 }4,167 2,945 5,221 5,976 5,211 5,484 4,698} 6 ,101 5,902 2 ,952 5,582 • 7 ,094 6 ,210 4,691• 4,353 5 ,478 4,859 5 ,451 5,058 5,910 3,216 4,908 5,925 5,157 5,617 5,375 6,040 2,935 5 ,425 • 6,936 6,128 4 ,766• 3,817 4,603 4 ,684 5,276 12.5 11.7}10.5 11.4 8.3 }10.2 6.6 12.0 12 .1 12.1 11.8 10.7}11.5 12 .1 7 .4 12 .3 6.4}9 .5 8.2 12 .5 12 .1 10.7 12.1}11.6 12.0 12.3 11.8 12.2 12.2 12.0 8.8 9.8 12.3 12.3 12 .1 12 .2 12 .2 12 .8 11.5 6.4 10.9 11.5 12.5 10.7 12 .2 10.0 11.7 12.1 9,180 8,925}7 ,841 8 ,839 4,894 }5 ,875 4,664 8,468 10,019 7,963 9,271 8,003}10,215 9,875 4,913 8,474 4,565}6,1 09 5,369 10,616 9,385 7,734 8,892} 7,715 8,009 8,869 7,857 7,972 10,814 10,157 5,084 7 ,323 11,202 6,528 10,845 11,265 10,853 12 ,500 6,842 4,408 7,431 8,474 9,359 7,250 9 ,751 5,351 8,778 7,598 •Not available. Sources: U.S. Department of.Commerce, Bur.eau of the <;ensus: Census of Population 1960 (Washington, D.C.: Government Printing Office, 19.61), Part C, General Social and Econo~1c Charactenstics, tables 73 and 76 ; Census ofPopulation 1970 (Washington, D.C.: Government Prmtmg Office, 1971), Part C, General Social and Economic Characteristics, tables 83, 89, and 90. to avoid the severe intrametropolitan cleavages afflicting almost every other SMSA in the nation. Texas annexation statutes have empowered every cen­tral city to extend its boundaries with only modest limits on the frequency and size of the annual increments. Through strategic application of this tool, the state's central cities have incorporated new suburban subdivisions into the enlarged city limits. Consequently, the propensity of afflu­ent, well-educated residents to settle at some distance from the central business district has not resulted in "white flight." Instead, as the geographical pattern of the new housing market has lured those with a high socioeconomic status to settle beyond the old city limits, the central cities have reached out, expanded their borders, and thereby avoided the serious problem of political fragmentation and jurisdictional polarization so common in the Northeast. During the 1950s, the decade marking the first wave of large-scale metropolitan growth in the Sunbelt, the central cities in Texas were frenetically annexing new territory. As a result, in many instances these cities offset pending popu­lation declines inside their outmoded political boundaries, and two-thirds of the central cities maintained a socioeco­nomic advantage over their unannexed suburbs. During the 1960s, however, the legislature restricted the unlimited power of central cities to annex new subdivisions and, in effect, dramatically reduced the rate and size of new territorial acquisitions. Not surprisingly, these new restric­tions also reduced the number of central cities having a socioeconomic advantage over their suburbs. Fortunately though, the trend towards greater intrametropolitan polari­zation may represent only a temporary phenomenon. Preliminary evidence suggests that as central cities adjusted Table 2 Socioeconomic Status of Selected Northeastern Cities and Suburbs, 1970 Median school years completed Median family income Ce ntral cities Central city Urban balance Central city Urban balance Baltimore 10.2 12 .1 8,815 11,816 Boston 12. l 12 .5 9,133 12,477 Buffalo 10.2 12 .2 8,804 11 ,453 Chicago 11.2 12.4 10,242 13,687 Cincinnati 11.2 12.2 8,894 11,669 Cleveland 10.7 12.4 9,107 12,985 Detroit 11.2 12 .3 10,045 13 ,371 Indianapolis 12 . I 12 .2 10,7 50 10,920 Milwaukee 12.0 12.4 10,262 12,916 Minneapolis St. Paul 12 .2}12 .2 12 .5 9,960} 10,544 12,732 New York 11.4 12 .3 9,682 1 3 ,811 Newark 10.0 12.3 7,735 13,577 Philadelphia 10.9 12.3 9,366 12,236 Pittsburgh 11.6 12 .2 8,800 10,132 Providence 10.9 11.2 8,430 9,907 St. Louis 9.6 12 .3 8,182 12,139 Toledo 12 .0 12.4 10,474 12,401 Youngstown 10.9 12.2 9,078 11,037 Source: U.S. Department of Commerce, Bureau of the Census, Cen­sus of Population 1970 (Washington, D.C.: Government Printing Office, 1971 ), Part C, General Social and Economic Characteris­tics, tables 83, 89, and 90. to the new annexation statutes (and in some instances devised ingenious strategies to offset some of the legisla­tively imposed restrictions), the pace of new annexations quickened appreciably. Thus, a judicious use of their annexation powers may enable many of the state's central cities to halt, and possibly even reverse, the drift toward greater intrametropolitan polarization. Clearly, present conditions in Texas metropolitan areas suggest that pessimism would be premature. Any percep­tible evidence of a Texas central city facing the extreme problems of many northeastern cities can be easily offset by the annexation practices permitted under Texas law. However, the opposition of suburban residents who face annexation by less affluent central cities, such as Clear Lake City's opposition to annexation by the city of Houston, and the potential restrictions on future annexations throughout Texas posed by the 1975 Voting Rights Act could nullify these advantages and reinforce the current drift toward greater intrametropolitan cleavages. Under these circum­stances, many central cities in Texas may well find them­selves involuntarily following the same path of heightened polarization and central city abandonment that has plagued many northeastern cities during the past thirty years. Even if Texas cities should successfully avoid the pitfalls and mistakes committed by their northeastern counter­parts, it is not clear that they will have solved some of the pressing social problems that have been the source of urban strife throughout the nation. In the Northeast, socioeco­nomic disparities and an inability to annex new territory have resulted in a political and jurisdictional separation b~­tween the rich and the poor, the well educated and the poorly educated. In Texas, by comparison, the rapid terri­torial expansion of the central cities has concentrated all socioeconomic groups into one political jurisdiction. Socioeconomic distress has not disappeared from Texas cities; rather it has been statistically camouflaged by the inability of wealthy residents to isolate themselves politi­cally in independent units of local government. Both in the Northeast and in Texas, the sources of large pockets of poverty are identical-individuals who have been displaced from their rural origins either in Latin America or the U.S. South and who have migrated to metropolitan centers in search of greater economic opportunity and social services. When these individuals first arrived in the Northeast, their concentration in urban settings made it easier to organize, maintain a spirit of political militancy, and place greater demands on local government for improved social benefits. In the Northeast, this political and social trauma is largely a historical curiosity, although its impact is still evident in the fiscal crisis and declining economic base of many of the central cities in that region. In Texas, the impact of large concentrations of poverty-stricken, poorly educated urban residents has not produced the same political and economic turmoil. Should these tensions emerge in the future, how­ever, the fact that intrametropolitan polarization is less pro­nounced in Texas than in the Northeast will be small conso­lation. The fact remains that significant social disparities do exist-only in Texas they appear within the same political jurisdiction. Texas facl Book by Charles P. Zlatkovich, Rita J. Wright, and Robert S. Moore, gives basic information on the economy of Texas, with particular attention to population trends, sources of income, major incfustries, and trading patterns. Small maps highlight special features of the 1978 pocket-size production. Statistics are current as of July. Paper $4. 00. ISBN 87755-227-4 Texas family Law (fifth edition), by Jack W. Ledbetter, was originally designed to supplement business law texts and is presently used as a text in courses in family planning, adult education, and general business law. The only up-to-date work in the area, Texas Family Law is written in layman's language. Mr. Ledbetter is general counsel for Lamar Savings in Austin. Paper $5.00. ISBN 87755-226-6 Texas residents add 5% sales tax. Bureau of Business Research The University of Texas at Austin Box 7459 Austin, Texas 78712 The Federal Economic Stimulus Package Two of the three major components of President Car­ter's Economic Stimulus Package (ESP) were casualties of the 95th Congress, which adjourned in October of 1978. Because the Anti-Recession Fiscal Assistance (ARFA) and Local Public Works (LPW) programs were discontinued, the five largest Texas cities-Dallas, El Paso, Fort Worth, Hous­ton, and San Antonio-lost revenue sources that had pro­vided them with a total of $32.4 million during fiscal 1978. While this amount is considerably below the $320.7 mil­lion in fiscal 1978 aid that these programs had provided to New York City, some impact on the budgets of these Texas cities is likely to be felt, especially if the programs are not reinstated this year. In the absence of federal funds, Texas cities face the alternatives of scaling down city services or increasing local taxes in order to maintain existing service levels. Despite being the most controversial component of the ESP, public service employment programs under the Com­prehensive Employment and Training Act (CETA) were extended relatively intact through fiscal year 1982. How­ever, the Carter administration has recently proposed reduc­ing the public service jobs component of CET A by about $535 million in fiscal year 1980, a move which would result in the elimination of approximately 158,000 jobs national­ly by the end of that fiscal year.1 All of the five largest Texas cities except Dallas would have been adversely affected had CETA been terminated last year. CETA was the least effective of the ESP programs in providing aid to cities most in need, although the 1978 legislation takes steps to improve future effectiveness. There are at least three major reasons why the 95th Con­gress failed to endorse ARFA, LPW, and other segments of the administration's urban policy .2 First, because a number of bills were introduced late in the session, insufficient time remained for detailed and informed hearings, debate, and Joseph E. Pluta is Research Economist, Texas State Comptroller's Office. The opinions expressed in this article are solely those of the author; the contents in no way reflect the position or policy ofthe Office ofthe Comptroller ofPublic Accounts. Effect on Large Texas Cities Joseph E. Pluta reflection. Second, programs calling for increased govern­ment spending were viewed with disfavor because of con­cern over both inflation and taxpayer sentiment. Third, some Sunbelt legislators from the South and West were re­luctant to support measures that would primarily benefit Snowbelt cities of the Northeast and Midwest. Nevertheless, a countercyclical revenue-sharing program, similar to ARFA, is under consideration in the ongoing 96th Congress but at substantially reduced funding levels. Extensive congressional battles are anticipated, and the outcome, at this writing, is uncertain. The LPW program, however, has been dropped from the Carter administra­tion's fiscal 1980 budget proposals. Although the precise impact cannot be measured at this time, reduced funding for both programs might especially affect Texas cities since recent congressional concern has been with ensuring that only the most needy cities receive federal aid. Most Texas cities do not appear to fall into that category. At this point, it is possible to present a synopsis of the background legislation and philosophy of ESP programs, focus on recent congressional activity, analyze the impact of ESP on Texas cities through September of 1978, deter­mine how effectively ESP programs have been directed to cities most in need, and estimate the amount of revenue needed from alternative local sources if various components of ESP are not reinstated and if cities choose to maintain existing public service levels. Unless otherwise noted, all figures cited in this article are taken from the U.S. Depart­ment of the Treasury, Report on the Fiscal Impact of the Economic Stimulus Package on 48 Large Urban Govern­ments, January 23, 1978. Background Legislation and Philosophy of Programs The ARF A program was intended as a temporary gen­eral purpose revenue supplement to local revenue sources when the latter declined because of sluggish economic con­ditions. The philosophy behind such a program was to maintain local spending efforts without local tax rate in- creases that might further restrict economic recovery. ARFA was authorized by Title II of the Local Public Works Employment Act of 1976 (PL 94-369) and extended through fiscal 1978 by Title VI of the Intergovernmental Anti-Recession Assistance Act of 1977 (PL 95-30). Discus­sion about extension of countercyclical aid last year cen­tered on revising the distribution formula and eligibility requirements so that future funding would be awarded only to the most fiscally distressed cities. Although the Senate passed such a bill (HR 2852), Congress adjourned before House floor action could be taken. Since the Senate version called for $500 million in local aid nationwide in fiscal 1979 and $340 million in fiscal 1980, some cities lost sub­stantial amounts of revenue: two-year losses of $85 .4 mil­lion in New York and $21.8 million in Chicago headed the list.3 Funding, therefore, has ceased (at least temporarily) for a program viewed as crucial to those cities most in need and to the success of the administration's urban policy. The LPW program was intended to stimulate the econo­my by initiating public projects that were not funded or approved for bond authority because of lack of local mon­ey. Under the Public Works Employment Act of 1977 (PL 95-28), the Commerce Department's Economic De­velopment Administration was responsible for allocating funds to states and localities for various capital improve­ments, such as school buildings, municipal offices, streets, and sewers. As part of his urban program announced in March 1978, President Carter proposed changing LPW from a capital-intensive to a labor-intensive program on grounds that the high cost of lumber and other building materials would make extension of the original program highly infla­tionary. Partly as a result of strong lobbying by the nation's construction industry, however, the House subcommittee considering the bill in 1978 proposed a compromise bill containing a mix of "hard" (capital-intensive) and "soft" (labor-intensive) public works. Strong opposition in the Senate eventually killed any LPW bill. Interestingly enough, one argument that emerged during Senate discussion of the measure was that the distribution formula for LPW funds discriminated against Texas. Be­cause of its size, Texas has many unemployed persons. Nevertheless, Texas has an overall unemployment rate that is below the national averar, and this fact limited the funds the state could receive. The CET A program was originally designed to reduce unemployment by creating temporary public service jobs until economic recovery generated additional employment in the private sector. CETA (PL 93-203) was approved De­cember 28, 1973, amended by the Emergency Jobs Unem­ployment Assistance Act twelve months later, extended in October 1976 (PL 94-444), and funded until September 1978 under the economic stimulus appropriation of May 1977. Title II authorized public service employment and other manpower services in geographic areas where unem­ployment rates were 6.5 percent or above. Title VI pro­vided employment slots in such community agencies as hospitals, schools, libraries, parks and recreation centers, public works, and police and fire departments. In recent years, CET A has been surrounded by controversy. Oppo­nents have contended that it is riddled with abuse and fraud , while advocates have argued that the program is es­sentially sound, that cities are in need of financial aid, and that steps have been and will be taken to correct abuses. Corrective action in the new bill (PL95-524) consists largely of increasing Labor Department authority to moni­tor the use of funds. The bill also contains special provi­sions to prevent local governments from substituting CETA jobs for employment they normally would have financed themselves. A ceiling is also placed on CET A wages. In answer to the charge that CET A was not effectively de­signed, the new measure limits participation to the eco­nomically disadvantaged and unemployed and places special emphasis on removing people from welfare roles by helping them learn skills. The final version of the bill provides for Table 1 Total and Per Capita ESP Allocations* (Total in millions of dollars; per capita in dollars) ARFA LPW CETA All programs Per Per Per Per City Total capita Tot al capita Total capit a Tot al capita Dallas 0 0 5.3 6.7 12.2 15.4 17 .5 22.1 El Paso 7.9 19.2 9.0 21.9 15 .4 37.5 32.3 78.6 Fort Worth 0.4 1.2 4.3 12.9 6.3 19.0 11.0 33.1 Houston 2.6 1.8 8.6 6.1 24.9 17.8 36.1 25.7 San Antonio 4.9 5.7 5.6 6.5 21.5 25.l 32.0 37.3 New York 245.3 33.6 289.3 39.7 298.0 40.9 832.6 114.2 Newark 17.2 52.4 29.0 88.4 27.1 82.6 73.3 223.4 Chicago 38.9 13.2 57 .1 19.4 103.9 35.3 199.9 67.9 Atlanta 5.1 12.7 9.8 24.5 28.3 70.9 43.2 108.l Tulsa 0.9 2.6 1.4 4.1 5 .I 15.1 7.4 21.8 Oklahoma City 1. 5 4 .0 6.4 17.3 45 .9 9.1 24.5 17.0 Phoenix 4.1 5.7 8.1 11.4 33.8 47.6 46.0 64.7 *From initial funding of each program through September 1978. Source: U.S. Department of the Treasury, Office of State and Local Finance, Report on the Fiscal Impact of the Economic Stimulus Package on 48 Large Urban Governments (Washington, D.C.: Government Printing Office, January 23, 1978). about 660,000 jobs nationwide and shifts the overall focus toward job training for the hard-core unemployed. Economic Stimulus Package Funding Local governments nationwide have received $15.8 bil­lion in ESP funds through September 1978. Approximately $3.2 billion of this amount has gone to the forty-eight larg­est urban governments. Of these forty-eight cities, New York has collected the largest amounts both in total ESP funds ($832.6 million) and in funds from each of the com­ponent programs ($245.3 million, ARFA; $289.3 million, LPW; $298.0 million, CETA). The smallest amounts in total funds ($7.4 million) and in CETA allocations ($5.1 million) went to Tulsa. Dallas received nothing from the ARF A pro­gram, and Louisville received the least amount ($0.6 mil­lion) under LPW. Of the five Texas cities, Fort Worth re­ceived the smallest amount in total ESP monies ($11.0 mil­lion) to rank forty-sixth among the forty-eight cities sur­veyed. The largest total ESP allocation to a Texas city went to Houston ($36. l million), which ranked twenty-sixth in this category. The ESP profile for each of the five Texas cities and for seven additional cities representing various regions of the country is presented in table 1. Per capita allocations are probably more meaningful for intercity comparisons than total dollar amounts are. The average ESP allocation per capita for the forty-eight largest urban centers was $83.33. Each of the five Texas cities re­ ceived less per capita than this figure . Nationally, the high­ est per capita amount was received by Newark ($223.40) and the smallest by Tulsa ($21.80). Dallas ranked forty­ seventh ($22.10) and Houston forty-fifth ($25.70). El Paso, which received the greatest per capita amount ($78.60) of the Texas cities, ranked twenty-first among the forty-eight cities. The importance of ESP allocations by region merits further attention. The U.S. Treasury Department classifies the forty-eight largest urban governments into five regions: North East, North Central, South East, South Central, and West. The South Central region includes the five Texas cities plus Oklahoma City, Tulsa, Kansas City, and St. Louis. The North East region received the largest shares of ARFA (40.4 percent), LPW (33.9 percent), and total ESP (29.9 percent) monies. The West region was allocated slightly more than the North East in CETA funds (23 .8 per­cent as opposed to 23.l percent). Of all the regions, how­ever, the South Central received by far the lowest portions in all categories (ARFA, 8.4 percent; LPW, 9.8 percent; CETA, 13.2 percent; and total ESP, 11.2 percent). A closer look at funding for each of the programs will reveal some of the reasons for relatively small allocations to the South Central region, including Texas, as well as the potential impact of both the programs and their possible termination upon Texas cities and other selected cities. ARFA Dallas did not receive ARF A funds because of low un­employment; if the local unemployment rate was less than 4.5 percent, no ARFA money was sent. Low unemploy­ment was also responsible for the low ARF A allocation to Fort Worth in fiscal 1978 and the absence of funds to that city for the first quarter of fiscal 1979. San Antonio and Fort Worth did not receive ARF A money in fiscal 1977 be­cause of the timing of their fiscal years. Funds allocated to these two cities during census fiscal year 1977 did not show up in their budgets until fiscal year 1978. Treasury Department estimates are available for the amount of money each city would have received during fis­cal year 1979 if funding levels under the old legislation had been extended. According to these estimates, termination Table 2 LPW Allocations by Fiscal Year Total allocations Total allocations to city governments to city areas• Round I (FY 1977) and Round II ( FY 1978) Ro und I and Round II City A mo unt (millions of dollars) Percentage of capital o utlay Percentage of long-term debt issue Number of projects Ro und I Round II A mount (millions of dollars) Number of projects Dallas El Paso Fort Worth Ho uston San Antonio 5.3 9.0 4.3 8.6 5.6 5.1 26.9 20 .0 2 .4 1. 3 12.8 166.7 20.4 5.5 2 .5 0 3 0 6 2 2 3 4 2 2 5.3 11.9 4 .3 14 .5 14.1 2 13 4 10 12 New Yor k 289.3 10.0 11.0 49 80 295.0 136 Newark 29.0 71.3 72.5 2 4 32.l 10 Chicago 57 .1 21.0 29 .6 7 16 66.8 33 Atlanta 9.8 6.5 9.8 1 1 14.1 5 Tulsa 1.4 2 .0 5.1 0 5 1.4 5 Oklahoma City 6.4 22. 5 15. 5 0 2 10 .2 7 Phoenix 8.1 7 .9 11.8 0 12 21.1 24 •includes all governmental units, not just the city government. Source: U.S. Department of the Treasury, Office of State and Local Finance, Report on the Fiscal Impact of the Economic Stimulus Package on 48 Large Urban Governments (Washington, D.C.: Government Printing Office, January 23, 1978). of ARFA will have some impact on El Paso and San Anto­nio ($3.4 and $2 .7 million in estimated lost revenue), rela­tively little effect on Houston ($1 .0 million in estimated lost revenue), and no immediate effect at all on either Dal­las or Fort Worth. In fiscal year 1979 and after, all other cities in the South Central region, except St. Louis, stand to lose relatively small amounts if ARFA is permanently terminated. In contrast, the estimated loss to New York City during fiscal year 1979 alone is a staggering $90.8 million . The precise allocation under prospective new legislation to each of the Texas cities cannot be determined at this time, but annual amounts would likely be less than the funding levels anticipated under the old legislation. El Paso and San Antonio have received rather large ARFA payments because of their high unemployment rates. Lower unemployment, therefore, especially in Texas cities, is the major reason for lower ARFA allocations to the South Central region. Yet relatively high unemploy­ment in El Paso (I0.7 percent in 1976) and San Antonio (8 .2 percent in 1976) has resulted in rather sizable ARFA payments to these cities. LPW LPW allocations were based on local unemployment, in­come levels, population, construction employment, and other related factors. The strength of the Texas economy in a number of these areas is a likely reason why LPW pay­ments to Texas cities were relatively low by national stan­dards. The $4.3 million total allocated to the Fort Worth area ranked forty-sixth among the forty-eight large urban areas nationwide; the $5 .3 million received by the Dallas area ranked forty-fourth nationally (see table 2). The $14.5 million received by the Houston area (the largest allocation to a Texas metropolitan area) ranked only twenty-ninth among the allocations to the forty-eight large cities. LPW allocations to cities as a percentage of their capital outlay and long-term debt issue may provide some insight into each city's financial profile. On the basis of existing data, however, it simply cannot be determined if LPW funds have resulted in new capital projects or have been merely an added source of funding for existing projects. What can be said is that in cities where the ratio of LPW funds to total capital outlay is high, the absence of such funds would have forced cities either to reduce capital ex­penditures or to finance their capital projects by such alternative means as bond issues. LPW funds certainly could have been used as a substitute for a city's normal borrowing program. Among Texas cities, El Paso and Fort Worth were much more dependent upon LPW funds for their capital projects than were Dallas, Houston, or San Antonio. The ratio of LPW funds to long-term debt issue was also greatest in El Paso and Fort Worth. Nationally, the local public works projects given greatest priority were streets, roads, and highways; water and sewer systems ; gymnasiums and swimming pools; and schools. These are the types of public projects into which a major portion of LPW funds have been channeled in various Texas cities. CETA Economic stimulus funds under the CET A program were intended to provide 725 ,000 jobs in the state and local sec- Table 3 CETA Allocations and Jobs Provided through September 1978 Total CETAjobs CETAjobs economic Jobs through CETA as percentage of as percentage of stimulus Titles II and VI • baseline total city City (millions of dollars) (in th ousands) work force jobs Dallas 12.2 1.3 9.8 8.9 El Paso 1 5 .4 1.7 48.6 32.7 Fort Worth 6.3 0 .7 15.9 13.7 Houston 24.9 2.7 18.9 15.9 San Antonio 21.5 2.3 20.7 17.2 New York 298.0 32.4 9.3 8.5 Newark 27.1 2.9 15.8 13.6 Chicago 103.9 11.3 23.2 18.8 Atlanta 28.3 3.1 34.8 25.8 Tulsa 5.1 0.6 18.2 15.4 Oklahoma City 9.1 1.0 26.3 20.8 Phoenix 33.8 3.7 50.0 33.3 *The Treasury Department derived the number of jobs by dividing total CETA allocations by average unit cost ($9,200). Source: U.S. Department of the Treasury, Office of State and Local Finance, Report on the Fiscal Impact ofthe Economic Stimulus Package on 48 Large Urban Governments (Washington, D.C.: Government Printing Office, January 23, 1978). tor nationwide. The 310,000 jobs funded before January 1977 are often referred to as sustainment allocations. After that date, funds labeled Carter stimulus allocations were made available to provide 415 ,000 additional jobs. Both of these components of the CET A program were scheduled to expire in September 1978 but have been renewed. Of the forty-eight large cities nationwide, only Tulsa ($5 .l million) received less in total CETA funds than Fort Worth ($6 .3 million). Houston, the largest Texas recipient ($24.9 mil­lion), was tied for sixteenth among the forty-eight large cities. New York ($298.0 million), Los Angeles ($118.5 million), and Chicago ($103 .9 million) were the top three recipients of economic stimulus funds. All data on CET A jobs and on the baseline work forces (full-time employees excluding CETA positions) in cities are Treasury Department estimates (see table 3). These estimates are not intended to indicate the actual number of CETA jobs a particular city has had and therefore would have lost had CETA funds expired. Precise data on the number of CETA jobs are not available for each of the forty-eight cities. The estimated number of jobs funded by CETA in large Texas cities has ranged from 700 in Fort Worth to 2,700 in Houston. The estimated 1,700 CETA jobs in El Paso appear to be far more important, however, when considered as a percentage of that city's baseline work force (48.6 percent) or total city jobs (32.7 percent). Both of these figures rank El Paso fifth nationally behind San Diego, Oakland, San Jose, and Phoenix. The potential for layoffs or reduced city services is certainly much greater in El Paso than it would be in other large Texas cities if CET A funds were to be re­moved. Some impact would also likely be felt in San Anto­nio, Houston, and Fort Worth, but the effect in Dallas would be rather small. In fact, the ratio of CET A jobs to total city jobs for the forty-eight large cities nationwide shows that only Nashville would be affected less than Dallas if CETA funds were eliminated. ESP Effectiveness and Revenue Alternatives The ESP was designed to stimulate cities most in need. In order to analyze the effectiveness of each of the compo­nent programs, the Treasury Department developed a fiscal­strain index. The method of measuring fiscal strain and the health of Texas cities according to this measure are dis­cussed in an earlier issue of Texas Business Review. 5 In re­sponse to fiscal need, ESP money has been reasonably well directed to the forty-eight cities with the possible exception of funds for the CET A program. A similar conclusion was reached in an August 1978 study by the Congressional Budget Office entitled City Needs and the Responsiveness of Federal Grants Programs. According to the Treasury, 53 percent of total ESP allocations to the forty-eight cities went to high-strain cities, 37 percent to those with mod­erate strain, and 10 percent to those with low strain. Ratios of high-to-moderate-strain and high-to-low-strain per capita allocations reveal ARFA as the most effective program and CETA as the least. High-strain cities received somewhat greater ARF A and LPW allocations per capita than moderate-strain and low-strain cities. The differences in CETA monies received per capita by high-strain, moderate­strain, and low-strain cities is only slight. Considering the tentative nature of federal ESP aid, cities are faced with reducing services or raising revenues from their own sources if ESP funds are terminated. Alter­natives involving reduced services will not be considered here since individual cities would be likely to slice or con­solidate existing programs in different ways, depending upon their perceptions of local priorities. Since revenue data are available, the option of raising city revenue can be discussed . The term city revenue is used here to represent revenue from the city's own sources, such as property taxes and city sales taxes. If ESP funding permanently ceases either in whole or in part and cities decide to maintain their existing levels of ser­vices, how much additional revenue would Texas cities be forced to generate? Bases on 1978 estimates, the termina­tion of all ESP funds would have had a particularly severe impact on El Paso, where ESP dollars were estimated to be 32 percent of city revenue (see table 4 ). El Paso ranks fourth among the forty-eight cities, trailing only Pittsburgh, Newark, and Buffalo. As a proportion of El Paso's city reve­nue, ARFA funds totaled 7 .2 percent, second nationally after Newark (7 .5 percent); LPW funds for El Paso were 8.6 percent of city revenue, ninth highest nationally; and CET A money made up 16.2 percent, making El Paso sixth nation­ally. While the reliance of San Antonio on CET A funds (12.7 percent of city revenue) is also relatively high (tenth nationally), the other Texas cities would have been affected to a lesser degree had total ESP funding ceased last year. Yet, even when reliance on ESP money is low relative to other cities, the termination of funds would mean that some services would have to be cut or some additional reve­nue would have to be generated. Assuming that measures similar to ARFA and LPW are not reinstated this year, the losses in Houston and in Dallas Table 4 Ratio of 1978 ESP Allocations to Adjusted Local Revenue (Percentage) City ARFA LPW CETA All programs Dallas 0 2.2 3.3 5.5 El Paso 7 .2 8.6 16.2 32 .0 Fort Worth 1.3 5.5 5.4 12.2 Houston 0.4 1.2 4.3 5.9 San Antonio 2.0 3.7 12.7 18.4 New York 2.0 2.8 3.0 7.8 Newark 7.5 15.2 16.6 39.3 Chicago 2.3 4.0 8.1 14.4 Atlanta 1.5 3.2 12.0 16.7 Tulsa 0.1 1.4 3.5 5.0 Oklahoma City 0.5 7.6 7.2 15.3 Phoenix 1.5 6.1 16.9 24.5 Source: U.S. Department of the Treasury, Office of State and Local Finance, Report on the Fiscal Impact of the Economic Stimulus Package on 48 Large Urban Governments (Washington, D.C.: Government Printing Office, January 23, 1978). (1.6 percent and 2.2 percent of city revenues) would be minimal; but revenue shortfalls in San Antonio and in Fort Worth (5.7 percent and 6.8 percent of city revenues) would be somewhat more noticeable. El Paso would lose an amount equivalent to 15 .8 percent of its city revenue, the fifth highest percentage nationally. A second important issue, then, is how much existing local tax rates would have to be increased to substitute for terminated ESP funds in each city. Given the tone of recent sessions of the Texas legislature as well as the nationwide "tax revolt," property tax increases are unlikely in Texas and, perhaps, anywhere else. Treasury Department data, however, may be valuable in estimating the amount of addi­tional revenue that would be needed from some source (alternative tax, state aid, or something similar) if Texas cities were to choose to maintain existing levels of services. If the entire ESP had been eliminated, property taxes per $100 of fair market value would have to be raised an average of sixty-five cents in high-strain cities, forty cents in moderate-strain cities, and twenty-four cents in low-strain cities to maintain existing expenditure levels. Among the forty-eight cities, the lowest increases would be in Houston and Nashville (seven cents each). The increases of twelve cents and nineteen cents in Dallas and Fort Worth would also rank low (forty-sixth and forty-third). The $ l .08 in­crease in El Paso would rank fifth in the nation. The rank­ing of Texas cities is similar in each of the ESP components. If measures similar to ARFA and LPW are not enacted in the current session of Congress, property tax increases necessary to maintain existing spending levels would be lowest in Nashville (one cent), second lowest in Houston (two cents), and third lowest in Dallas, Tulsa, and Memphis (five cents). The fifty-three cent increase in El Paso would be the fifth highest nationally. Notes 1. Democratic Study Group, U.S. House of Representatives, "The Administration's Jobs Budget," Special Report, no. 96-5 (Mar. 9, 1979): 6. 2. Congressional Quarterly, "Summary-95th Congress: No Sweep­ing Changes," Weekly Report 36, no. 42 (Oct. 21, 1978): 3013. 3. Congressional Quarterly, "Countercyclical Aid Bill Dies in Ad­journment Crush," Weekly Report 36, no. 42 (Oct. 21, 1978): 3038. 4. Congressional Quarterly, "House Unit Expands Carter Public Works Plan," Weekly Report 36, no. 35 (Sept. 2, 1978): 2394. Sixty-five percent of LPW money was allocated according to the number of unemployed people in each state, while 35 percent was distributed according to the percentage of unemployed. States, such as Texas, with unemployment rates below the national average did not qualify for the 35 percent portion. 5. Joseph E. Pluta, "Urban Fiscal Strain and the Health of Large Texas Cities," Texas Business Review, January-February 1979, pp. 8-12. ATLAS OF Box 7459, University Station Austin, Texas 78712 Manufacturing in Texas Four circumstances cause apprehension about the effect of a national recession on Texas manufacturing: (1) manu­facturing is one of the most volatile segments of the U.S. economy, (2) durable-goods manufacturing is the most volatile part of manufacturing, (3) manufacturing is in­creasing in importance to Texas relative to its importance to the United States, and ( 4) durable-goods manufacturing is the fastest-growing part of Texas manufacturing. As Evan Anderson and Lorna Monti pointed out in separate articles in the January-February 1979 issue of Texas Business Review, the growth of manufacturing has likely made the Texas economy more susceptible to fluctuations in the national economy than it has been in the past. A study supported in part by the Bureau of Business Re­search at the University of Texas at Austin was designed to measure the extent to which manufacturing investment and employment in Texas respond to changes in the level of national economic activity for the five largest durable-goods industries (primary metals, fabricated metals, nonelectrical machinery, electrical equipment, and transportation equip­ment) and three of the largest nondurable-goods industries (food, chemicals, and petroleum products). Because of the diversity and complexity of the many factors that impinge Lowell Dworin iB A18i1tant ProfeBSor ofAccounting, University of TexaB at Au1tin. How Recession Proof? Lowell Dworin upon the decision to invest in plant and equipment and to hire additional workers, any analytical description will capture only a limited portion of the many determinants of manufacturing growth within a state. By focusing on the impact of fluctuations in national economic activity, as measured by real gross national product (GNP), the Bureau of Business Research study shows that manufacturing employment in Texas is even more susceptible to a down­turn in the nation's economy than national manufac­turing employment, while manufacturing investment in Texas is at least as susceptible as national manufacturing investment to an economic downturn. Because the role of manufacturing is still somewhat smaller in the state econ­omy (18 percent of total 1975 employment) than in the national economy (24 percent) and because such fluctua­tions are masked by the general growth of manufacturing in the state, this susceptibility is not readily apparent. Never­theless, for better or worse, the state's manufacturing sector is tied to the health of the nation's economy. In this study net investment and changes in employment in manufacturing were related to changes in real GNP. Changes in real GNP are assumed to produce altered ex­pectations of future real GNP and thus future demand, which a manufacturer may meet by altering the investment in plant and equipment, the level of employment, or both. The analysis also assumes that investment in plant and equipment may always be substituted for increased employ­ment,1 although the extent to which the balance between investment and employment shifts in response to the cost of labor (including the indirect costs) relative to the cost of capital tends to differ among the various industrial sec­tors. The percentage change in gross plant and equipment and the percentage change in employment from 1961 to 1976 for the largest firms in each of the industrial sectors studied were used to estimate the extent to which a change in anti­cipated real GNP influences future demand and the extent to which an anticipated change in the cost of labor relative are thus assumed to be more quickly reflected in the behavior of employment than in investment. Once the relative importance of changes in real GNP and in the labor-capital ratio in each industry were obtained from an examination of the largest firms within the indus­try (which do most of the investing and employ the greatest numbers of workers), the specific characteristics of the in­vestment and employment response in Texas were obtained from an analysis of the percentage change in gross plant and equipment and the percentage change in employment in Texas. For this purpose an estimate of gross plant and equipment in each Texas industrial sector from 1957 to If a mild and short recession were to occur, employment in the durable-goods industries in Texas would be affected much more than employment in the nondurable-goods industries. to the cost of capital influences the desired labor-capital 197 6 was constructed by cumulating gross investment data ratio in each industry. Because of the greater irreversibility and adjusting for estimated retirements. of the investment decision process and the greater time re­quired to implement investment decisions, the change in anticipated real GNP and in the labor-capital ratio to which Employment Response to a Recession investment responds was assumed to be based on a three­year moving average of past changes, whereas the change in If a mild and short recession (characterized by a l anticipated real GNP and in the labor-capital ratio to which percent decrease in real GNP in the first year followed by a employment responds was assumed equal to the current l percent increase in real GNP in the next year) were to oc­changes. Short-term fluctuations in real GNP and wage rates cur, employment in the durable-goods industries in Texas Table 1 Effect of Mild Recession on Industrial Employment in Texas (Percentage change)* Year 1 Year 2 Year 3 Year 4 (1 percent reduction ( 1 percent increase (No change (No change Industry in real GNP) in real GNP) in real GNP) in real GNP) Primary metals -0.60 0.56 0.04 0.00 Fabricated metals -1.37 2 .34 -1.66 1.17 Machinery (except electrical) -1.22 1.65 -0.58 0 .20 Electrical equipment and supplies -3.47 3.57 -0.11 0.00 Transportation equipment -1.72 0 .52 0 .36 0.25 Food and kindred products -0 .13 0.09 0.03 0.01 Chemicals and allied products -0.56 0 .58 -0.02 0.00 Petroleum and coal products -0 .14 0.09 0.03 0.01 Texas aggregate•• -1.16 1.21 -0.28 0.22 U.S. aggregate•• -0 .84 0.67 0 .18 0.05 *These percentage changes in employment are a result of the assumed recession only; they will in general be superimposed on the responses generated from changes in real GNP before the initial recession year and from changes in GNP occurring after the recovery year, as well as from the effects of other factors. •*The aggregate Texas response is the average response for all the examined industries and is weighted in proportion to the number of em· ployees in each industry in Texas in 1975. This figure more closely reflects the existing mix of industries in Texas than does the simple average. The aggregate U.S. response is obtained from an analysis of the national employment response for the same industries weighted in proportion to 1975 national employment. would be affected much more than employment in the nondurable-goods industries, with the electrical equipment and supplies industry most responsive and food and kindred products and petroleum and coal products least responsive (see table 1). Moreover, the fabricated metals, machinery (except electrical), electrical equipment and supplies, and chemicals and allied products industries appear to over­respond; their adjustment would be oscillatory rather than smooth.2 The effects of a deeper national recession would be proportionally greater; the response pattern would, however, be altered if a longer recession occurred. These results clearly support Anderson's concern that the Texas economy is becoming more susceptible to fluctu­ations, at least with respect to employment. A 1 percent re­duction in real national income results in a 1.16 percent re­duction in manufacturing employment in Texas, but only a 0.84 reduction for the nation. The recovery of Texas manufacturing is, however, also magnified, and the effects of the recession persist for several years; they die out more rapidly for the nation. Investment Response to a Recession The decrease in investment in Texas in response to a downturn in the national economy is less severe than the reduction in employment. It is only slightly greater than the national reduction in investment (see table 2). At first glance it may appear surprising that Texas manu­ facturing investment, which has been increasing much more rapidly in the state than it has nationally, is only slightly more responsive to real GNP than manufacturing invest­ ment for the nation as a whole. There appear to be certain locational factors responsible for much of this growth of in­ vestment in Texas that are not directly related to real GNP. In fact, the three top investing industries in Texas (petro­leum and coal products, chemicals and allied products, and primary metals) grew only moderately in output during the l 970s.3 While investment in these industries is thus less closely tied to aggregate national economic activity than might be supposed, it is, of course, possible that these special factors (such as the availability and price of energy and other resources) will decline in importance and, in so doing, create their own fluctuations in investment in Texas. The decrease in investment in Texas in response to a recession would be less severe than the reduction in employment. Despite the long-term growth in manufacturing (which has been slowing), fluctuations in real GNP do appear to have been reflected in Texas manufacturing employment and investment, particularly in durable-goods industries. Notes 1. A Cobb-Douglas production function was used in this analysis. 2. To some extent this behavior may be an artifact of the estimation procedures; for fabricated metals and machinery (except electrical), a statistically satisfactory expression was obtained only by including as an additional factor the ratio of Texas personal income to nation­al personal income. 3. Art Ekholm, "Manufacturing Investment Booms in Texas," Voice of the Federal Reserve Bank of Dallas, November 1978, p. 11. Industry Table 2 Effect of Mild Recession on Investment in Industrial Gross Plant and Equipment in Texas (Percentage change) Year 2 Year 3 Year 4 ( 1 percent reduction (1 percent increase (No change (No change in real GNP) in real GNP) in real GNP) in real GNP) Year 1 • Primary metals Fabricated metals 0 0 -0 .37 -0.6 8 -0 .17 -0 .5 8 0.26 0. 19 Machinery (except electrical) 0 -0 .88 0 .04 0.88 Electrical equipment and supplies Transportatio n equipment Food and kindred products 0 0 0 -0 .98 -0 .2 3 -0 .0 8 -0 .11 -0.17 -0.02 0 .97 0.10 0 .07 Chemicals and allied products 0 -0 .20 -0. 14 0.1 0 Petroleum and coal products Texas aggregate•• U.S. aggregate .. 0 0 0 -0 .05 -0 .25 -0 .23 -0.05 -0 .11 -0.13 0.00 0 .16 0 .15 •No investment response is shown for year 1 because it is assumed that inves tment responds only to changes in real GNP during previous years; that this is a valid assumption may be seen by examining the pattern of investment in Texas described by Art Ekholm in the Novem­ber 1978 issue of Voice of the Federal Reserve Bank of Dallas . ..The Texas and U.S. aggregates are obt ained in a manner similar to that used in table 1 with weighting in proportion to the gross plant and equipment in each industry. The Industrial Screening Matrix Deborah G. Jones Douglas J. Manifold Jean S. Spencer In the highly competitive interregional search for new industries, development agencies need a systematic ap­proach for marketing their regions. The Industrial Screening Matrix was developed for the North Texas Commission, which serves eleven counties with 161 communities in the Dallas-Fort Worth area, and is designed to identify indus­tries that an agency might seek. The matrix-building process incorporates goals from several segments of the develop­ment community and gives developers insights into the re­gion that improve the presentation of the region to indus­trial prospects. The matrix process can be applied to any region served by a development agency. Description of the Matrix A matrix is a two-way table arranged to show relations between the variables in the columns and the items in the rows (see table 1 ). In the Industrial Screening Matrix, the columns constitute desirable traits for a local industry and the rows are the industries to be considered. Desired traits are determined by considering community wishes about industries to be recruited and advantages (such as conve­nient transportation or a skilled labor force) that the com­munity offers prospective industries. Each trait is given a weight that indicates its desirability. If an industry does not have the trait in question or does not meet the cut-off point established for that trait, nothing is added to the in­dustry's score for that trait. Deborah G. Jones, Douglas J. Manifold, and Jean S. Spencer are re­search associates with the Southwest Center for Economic and Com­munity Development, University of Texas at Dallas. For a more de­tailed report of their findings, see Southwest Center for Economic and Community Development, Development of a Dallas/Fort Worth Metroplex Target Industry Identification Program (Richardson : Uni­versity of Texas at Dallas, 19 78). The traits of each industry are marked by an X at the intersection of the row and each column. Scores are devel­oped for each industry by moving across the rows and add­ing the weights for all the traits represented by the col­umns. The ten criteria include such variables as average wages, need for rail transportation, type of labor employed, national market potential, natural gas consumption, and standards on water discharge quality. According to the weights established by the North Texas Commission for the Dallas-Fort Worth area, eighteen of the top twenty in­dustrial prospects are manufacturers of nonelectrical ma­chinery (see table 2). Other development agencies can follow a similar process to determine the most attractive industries for their areas. The process of developing a matrix involves five steps: or­ganizations in the region establish goals for industry charac­teristics ; organizations or consultants list the characteristics of the region; weights are assigned to those characteristics the region offers and wants; the matrix is constructed using the characteristics of the industries; and scores are assigned to industries. Techniques behind the Matrix The North Texas Commission study used five ap­proaches to classify the characteristics of industries and of the region: judgment of the development community, in­dustrial profile, shift-share analysis, location quotient, and industrial-linkage analysis. Another agency could choose either more or less rigorous analysis, depending on the complexity of the region and the resources available. Judgment of the Development Community Two phases of the screening process require the exercise of judgment by the development community: selecting lo- a:: > Table 1 ~ c:: Dallas-Fort Worth Screening Matrix (Samplel z tn -\0 -...) \0 Sources: (1) National Science Foundation, National Science Board, Science Indicators 1976 (Washington, D.C.: Government Printing Office, 1977); and U.S. Department of Com­merce, Industry and Trade Administration, 1978 U.S. Industrial Outlook (Washington, D.C.: Government Printing Office, 1978). (2) U.S. Department of Commerce, Bureau of the Census, Annual Survey of Manufactures 1976: General Statistics for Industry Groups and Industries (Washington, D.C.: Government Printing Office, 1978). (3) U.S. Depart­ment of Commerce, Bureau of the Census, Census of Transportation, 1972: Commodity Transportation Survey (Washington, D.C.: Government Printing Office, 1976). (4) Walter E. Mullendore, Arthur L. Ekholm, and Paul M. Hayashi, An Input-Output Model of the North Central Region of Texas (Arlington, Texas: University of Texas at Arlington, 1972). (5) U.S. Department of Commerce, Bureau of the Census, Census of Population, 1970: Occupation by Industry (Washington, D.C.: Government Printing Office, 1972). (6) U.S. Department of Commerce, Bureau of the Census, County Business Patterns 1975, Texas and U.S. Summary (Washington, D.C.: Government Printing Office, 1977). (7) U.S. Department of Commerce, Bureau of the Census, Annual Survey of Manufactures 1975, Fuels and Electric Energy Consumed (Washington, D.C. : Government Printing Office, 1977). (8) Same as source (2). (9) U.S. Department of Commerce, Bureau of the Census, Census ofManufactures 1972: Water Use in Manufacturing (Washington, D.C.: Government Printing Office, 1975). (10) U.S. Department of Labor, Bureau of Labor Statistics, Employment and Earnings, 1973 and 1978 issues; and U.S. Department of Labor, ~ Bureau of Labor Statistics, unpublished data, December 1977. cational criteria that will reflect the growth patterns of the region and ranking the criteria according to a set of priori­ties or goals upon which the development community is agreed. By the time developers have conducted a thorough regional analysis, they will have a factual understanding of their region that will allow them to choose a correct set of criteria and a weighting scheme for those criteria. of the economy but also to the future. A summary of the locational criteria and their associated weights and cut-off designations is presented in table 3. Some of the locational criteria were taken from a study on industrial development in Missouri.1 Such other criteria as water-discharge quality, high level of technology, and energy consumption were added because they were felt to be particularly important According to the weights established by the North Texas Commission for the Dallas-Fort Worth area, eighteen of the top twenty industrial prospects are manufacturers of nonelectrical machinery. Our consulting group was aided in the critical steps of selecting and weighting variables by an advisory committee composed of representatives of chambers of commerce, councils of governments, the city of Dallas, the Trinity River Authority, and the business community. At this stage of the process, advice from the community should be en­couraged. The size, structure, and ultimate use of the ma­trix will be determined at this stage. In the study for the Dallas-Fort Worth area, certain fac­tors were chosen as important not only to the present state Table 2 Twenty Most Attractive Industries SIC Industry Score• 3549 Metal wor king machinery, not elsewhere classifie d 9 1 3534 Elevators and moving stairways 89 353 5 Conveyors and conveying eq uipment 89 3555 Printing trades machinery 89 35 74 Calculating and accounting mac hines 89 3537 Industrial trucks and tract ors 88 3545 Machine tool accessories 88 3589 Service industry machinery, no t elsewhere classified 88 3533 Oil field machinery 86 3559 Special industry machinery, not elsewhere classified 86 3561 Pum ps and pumping equipment 86 3564 Blowers and fans 86 3569 Ge neral ind ustrial machinery, not elsewhere classified 86 3612 Transformers 86 3542 Machine t ools, metal forming types 85 3544 Special dies, t ools, jigs, and fix tures 85 3551 Food prod ucts machinery 85 3694 Engi ne electrical eq uipment 85 3 532 Mining machinery 83 3541 Machine tools, metal cutting types 83 *Out of a possible 98 points. Source: Southwest Center for Economic and Community Develop­ment, Development of a Dallas/Fort Worth Metrop/ex Target Industry Identification Program (Richardson: University of Texas at Dallas, May 197 8). to the future industrial development of the Dallas-Fort Worth area. The high-technology variable was the most difficult to quantify, but it was a priority item for the advisory committee. The final weighting scheme resulted from a survey of the advisory committee members in which several contrasting viewpoints were expressed. High technology was consistent­ly rated high by all members; next in importance were mar­ket potential and air transportation. Average and above­average wages were rated high by business leaders and re­gional developers. Suppliers in the region and buyers out of the region were important to everyone. Unskilled labor was important to some of the members but not to regional and business leaders. Everyone favored industries that hire more professional and technical people. Response toward the energy-consumption variable was uneven, and the water-discharge quality seemed most important to the business leaders. Industrial Profile The first step behind an analysis of a region's potential attractiveness for industry is to determine the area's exist­ing strengths and weaknesses. 2 An industrial profile details the structure and change in the regional economy and pro­vides basic information about which industries are presently growing and which are likely to expand in the future. Based on employment data at the SIC level of four digits, the Dallas-Fort Worth profile included employment change over three periods: 1970-1972, 1972-1975, and 1970­1975. Other indicators of growth can be used, but employ­ment figures are readily obtainable and reasonably com­plete at such a detailed level. The industrial profile produced a list of 234 manufac­turing industries (at the SIC level of four digits) that were analyzed in the matrix. The profile also contributed indi­rectly to three criteria: labor intensity (variable 8 in table 1 ), competitiveness (variable 10), and market potential (variable 6). TEXAS BUSINESS REVIEW Several sectors were found to be important contributors to growth in the Dallas-Fort Worth economy. The most active industry in the area has been manufacturing, with the retail and services sectors also showing significant growth. Trade employment has grown consistently since 1970, re­flecting growth in population. Showing the same general trend are finance, insurance, and real estate; transportation ; and public utilities. Construction employment, on the other hand, has declined. Shift-Share Analysis Shift-share analysis was used to determine the share of a regional economy in the total national economy. Changes in the Dallas-Fort Worth industrial structure were exam­ined over the period from 1972 to 1977. Because a region's industrial structure is dynamic, a shift-share analysis is most useful in monitoring the specific movement of the economy over a short period of time. Three components were assessed: the national component, the industrial mix, and the competitive component. 3 The competitive component was used to derive the com­ petitiveness variable in the matrix (variable 1 O). This vari­ able is a competitive quotient (actual change in regional em­ ployment divided by the competitive component), which is an indicator of the region's comparative advantage for an industry. An industry received a matrix score for competi­ tiveness if its competitive quotient was higher than the quotient for total manufacturing employment. Shift-share analysis demonstrated that the Dallas-Fort Worth area is attracting industries that are growing slowly nationally but faster regionally. location Quotient Location quotients identify a region's basic industries and indicate their potential for expansion. The location quotient is the ratio of actual employment to the employ­ment that would be needed to meet an area's requirements for goods and services in an industry. When the percentage of national employment in an industry is used as a base, a quotient of one means industries are fulfilling the needs of an average area and a quotient greater than one indicates support, or basic, industries. According to export base theory, basic industries sell a significant amount of their production to other regions and support the economic base of their area. Both shift-share analysis and location quo­tients can demonstrate the degree to which the regional economy is specialized or diversified. The location quotient was used as a rough gauge for mar­ket potential (variable 6 in table l ). Location quotients at the level of four SIC digits were computed for 1975 from the industrial profile. An industry with a location quotient of less than 1.2 was considered to have a regional market potential, while an industry with a quotient of 1.2 or more was considered to have a national market potential. Results for the Dallas-Fort Worth region suggest that in 1977 SIC categories 23 (apparel and other textile prod­ucts), 27 (printing and publishing), 36 (electrical equipment and supplies), and 37 (transportation equipment) were the Table 3 Screening-Matrix Variables and Weights Weight Locational criteria (0-1 O) Cut-off point (I) High level of technology (2) National wage levels Above average Average Below average (3) National transportation Rail Motor carrier Air (4) Interindustry linkages Backward in region Forward in region Backward out of region Forward out of region (5) National labor orientation Professional Craftsmen Laborer (6) Market potential Regional National (7) Energy Electric Natural gas (8) Labor intensity (9) Water-discharge quality (I 0) Competitiveness 9 High ratio of research and development expenditures to sales 5 Greater than or equal to $ 5.70 per hour 8 $5.17-$5 .69 per hour 6 Less than or equal to $5 .16 per hour 4 31 .7 percent of industry shipments distributed 8 18.3 percent of industry shipments distributed 10 0 .1 percent of industry shipments distributed 8 Greater than 30 percent of purchases within region 8 Greater than 30 percent of sales within region 1 Less than 30 percent of purchases out of region 6 Less than 30 percent of sales out of region 9 I 0.1 percent of industry labor requirements 7 19.6 percent of industry labor requirements 3 4.6 percent of industry tabor requirements 5 Less than 1.2 location quotient 8 Greater than 1.2 location quotient 8 Greater than 5,690 kilowatt-hours 7 Greater than 33.5 cubic feet 2 Less than or equal to 39.187 value added per production worker 5 Less than 0.435 ratio of treated water to total water discharge 3 Greater than 0.259 competitive quotient Source: Southwest Center for Economic and Community Development, Development ofa Dallas/Fort Worth Metroplex Target Industry Identi­fication Program (Richardson: University of Texas at Dallas, May 1978). basic industries significantly involved in selling products outside the region. The area also specializes in trucking and warehousing, wholesale and retail trade, and finance, insur­ance, and real estate. Industrial-Linkage Analysis An industrial-linkage analysis is used to uncover what supply and marketing ties a region has to the national econ­omy and to explore the possibilities for substituting input materials. Backward linkage indicates where an industry buys materials; forward linkage indicates where it sells products. Intraregional backward and forward linkages can be examined through the use of an input-output table.4 In the Industrial Screening Matrix, industries are evalu­ated on whether they have backward or forward linkages with existing regional industries (variable 4 in table 1). While an industry would find it profitable to have strong forward and backward linkages in the region, the local economy benefits more from forward linkages outside the region, since sales elsewhere increase the flow of money into the region. The screening process allows developers to eliminate marginal industries and Jocus on industries that have the best potential jor the area. The linkage analysis revealed that Dallas-Fort Worth has a relatively open economy. Sales of products to other re­gions and purchases of materials from other regions are brought about by a high level of specialization. Because of a dependence on the traditional manufacturing belt, the potential for local production of input materials is good. The Industrial Screening Matrix uses the list of suitable candidates that the industrial profile and shift-share analysis have provided and incorporates the locational considera­tions of the industrial linkages and location quotients. While each technique reveals significant information about the region, none is intended to stand alone. Together, how­ever, these techniques provide an accurate interpretation of a region's economy. Further Applications of the Matrix While one type of screening matrix has been presented here, there are several variations that might make it better suited to a different area. Another study has been modeled to meet the needs of a variety of Missouri communities of different sizes. It necessarily involved more steps in the actual screening process than did the Dallas-Fort Worth study. For example, developers in each community first needed to evaluate the area's attributes as a feasible site for plant locations. Because some communities lacked essential ingredients (for example, railway transportation or waste treatment facilities) for attracting specific industries, devel­opers in these areas would automatically eliminate some categories of industries as impracticable for attraction. This bridging of community attributes and industry needs is an essential first step for smaller communities but was not needed in studying Dallas-Fort Worth because the region can accommodate many different kinds of industries and because developers are aware of the attractions of the area. Community Strategy Formulating a development strategy for any community requires several important considerations. First, the scale of the project-the size or complexity of the area being developed-must be determined. The screening matrix can accommodate an entire state, a subregion of a state, a small city, or a cluster of cities. Second, the type of industry needed for the area's growth-whether manufacturing, retail, wholesale, or service-must be evaluated. This consideration is directly linked to the locational criteria that are used in the matrix. If developers are to concentrate on attracting service indus­tries to the area, they cannot use locational criteria that apply only to manufacturing activities. The industry re­quirements will be different for each sector. Also, an appropriate weighting scheme must be devised. The difficulty of this process depends on the time frame for the screening, the number and types of participants in­volved in the weighting, and the objectives of the devel­opers. Some experimentation with different weights may prove helpful. Finally, in this period of competitive strate­gies for development, it becomes increasingly difficult for developers to rule out industries for subjective reasons. The object of the screening process is to eliminate as many of the marginal industries as possible and focus on the indus­tries that represent the best potential for attraction. If used in this way, the Industrial Screening Matrix can be a useful tool in guiding developers toward directed goals of growth. Notes 1. Charles W. Minshall, John D. Miller, and Allen White, Final Re­port on Development ofa Community Level Target Industry Identi­fication Program to the State of Missouri, Division of Commerce and Industrial Development (Columbus, Ohio: Batelle.Columbus Laboratories, 1976). 2. For a useful model, see Lorna Monti, "Capsule Economic Analy­sis: How to Design an Economic Profile of Your Area," Texas Busi­ness Review, May 1976, pp. 104-9. 3. For a complete description of these components, see Walter E Mullendore and Lawrence F. Ziegler, "Regional Patterns of Indus­trial Growth in the Planning Regions of Texas," Texas Business Re· view, March 1978, pp. 53-57. 4. Walter E. Mullendore, Arthur L. Ekholm, and Paul M. Hayashi, An Input-Output Model of the North Central Region of Texas (Ar· lington: University of Texas at Arlington, 1972). Longview-Marshall: An Industrial Center The Longview-Marshall area has been a center of indus­trial activity for more than a century. Marshall was one of the largest and wealthiest cities in Texas before and during the Civil War and became both the seat of civil authority for the Confederacy west of the Mississippi after the fall of Vicksburg and an industrial center producing clothing, harnesses, ammunition, and other items. Longview, now much larger than Marshall, was not established until 1870 when surveyors for the Texas and Pacific Railway remarked on the long view from a high spot in the area. Throughout its history, natural resources, location, and transportation have enhanced the Longview-Marshall area as an industrial center. Energy resources have been particu­larly important for the past fifty years. Water is plentiful by Texas standards. Marshall and Longview were both impor­tant railroad communities from the early days, and Inter­state 20, one of the major southern transcontinental routes, passes through the area. Longview is 125 miles east of Dallas, 230 miles north of Houston, and 61 miles west of Shreveport. Population Growth The Longview-Marshall SMSA consists of two counties­Gregg and Harrison. The most recent official estimate by the U.S. Bureau of the Census is that 127 ,900 people lived in the area at midyear 1976, with about 64 percent of the population in Gregg County and 36 percent in Harrison County. The census estimate indicated that the area popula­tion had increased by about 5 .9 percent between 1970 and 1976, a growth rate slightly above the overall national population growth rate but well below the 11.5 percent gain in Texas. The Census Bureau attributes about 65 per­cent of the Longview-Marshall population growth to natural increase, or the excess of births over deaths, and the other 35 percent to net migration into the area. Net migration accounted for about 42 percent of the state's population growth during the same period, according to Census Bureau estimates. There are indications that the area growth rate has ac­celerated since mid-1976. A comparison of the rate of in­crease in motor vehicle registrations through the first quar­ter of 1978 indicates that only six of the twenty-five Texas Oiules P. Zlatkovich is Research Associate, Bureau ofBusiness Re­Bearch. MAY-JUNE 1979 Charles P. Zlatkovich metropolitan areas have exceeded Longview-Marshall in per­centage gain in vehicle registrations. By contrast, Longview­Marshall ranked nineteenth, or sixth from the bottom, in rate of population growth among the Texas metropolitan areas at mid-1976. There is a strong correlation between population and motor vehicle registration totals. Longview was first designated as a metropolitan area in August 1975 and is the most recent Texas area to achieve official metropolitan status. The official name of the metro­politan area was changed to Longview-Marshall in Decem­ber 1978. The Longview-Marshall area is adjacent to two other metropolitan areas: Shreveport, which is much larger with a population of about 350,000, and Tyler, which is slightly smaller. Employment Structure and Sources of Income There are sharp differences between the economy of the Longview-Marshall area and the economy of the state. Longview-Marshall is much more involved in manufactur­ing, and the direct contribution of government to the local economy is much lower than it is across the state. Manufacturing is the dominant sector in the Longview­Marshall economy. Almost 32 percent of nonagricultural civilian employment in the area is concentrated in the man­ufacturing sector, while about 18 percent of corresponding Nonagricultural Civilian Payroll Employment Percentages Longview-Marshall SMSA, Texas, and United States 1977 Annual Average Longview-Marshall United Category SMSA Texas States Mining 4.1 3.2 1.0 Construction 6.6 7.0 4.7 Manufacturing 31.8 18.4 23.8 Transportation , communication , and public utilities 6.1 6.3 5.6 Trade 22.6 24.6 22.3 Finance, insurance, and real estate 3.5 5.6 5.5 Services 14.1 17.2 18.7 Government 11.2 17.7 18.5 Sources: Data for Longview-Marshall SMSA and Texas obtained from Economic Research and Analysis Department, Texas Employment Commission; U.S. data obtained from Employment and Earnings (Washington , D.C.: U.S . Department of Labor, Bureau of Labor Statistics, A pril 19 77-M arch 197 8) . 103 statewide employment and 24 percent of national employ­ment is in manufacturing. More inclusive personal income statistics present a similar picture. Manufacturing accounts for more than 28 percent of Longview-Marshall personal in­come, l 5 percent of state personal income, and just under 20 percent of national personal income. The only Texas metropolitan area with a higher percentage of personal in­come derived from manufacturing is Beaumont-Port Arthur­Orange, whose residents derive about 33 percent of their in­comes from manufacturing earnings. With the manufacturing sector contributing a portion of area income that is 87 percent greater than the statewide contribution of that sector, the relative contribution of most of the other sectors would be expected to be below average. On the contrary, the mining sector also makes a notably larger contribution to Longview-Marshall income than it does statewide, and the contribution of the trans­portation, communication, and public utilities sector to the local economy is well above the state average. The economic activities that contribute notably less to the economy of the area than they do across the state or the nation are agriculture and the federal government. Agriculture was almost insignificant as a source of income in the Longview-Marshall area in 1976, although its contri­bution had been larger in earlier years. The total contribu­tion of federal earnings, both military and civilian, is less than 22 percent of the relative contribution of federal earnings to state income. State and local government is also a smaller contributor to personal income in the area than in Percentage of Personal Income by Major Sources Longview-Marshall SMSA, Texas, and United States, 1976 Longview- Marshall United Source SMSA Texas States Agriculture 0 .02 1.55 1.82 Mining 5.35 3.50 1.11 Construction 5.64 6.22 4 .37 Manufacturing 28.67 15.32 19.74 Transportation, communication, and public utilities 6 .58 6.08 5.69 Wholesale and retail trade 14.57 15.42 13.10 Finance, insurance, and real estate 2 .40 3.95 4 .01 Services 10.66 11.92 12.52 Other industries 0 .26 0 .26 0 .27 Total private labor and proprietor income 74.15 64.22 62 .65 Federal civilian 1.01 3.20 3.28 Federal military 0.15 2 .17 1.34 State and local 7.64 8.05 8.92 Total government earnings 8.80 13.42 13.53 Total labor and proprietor income (place of work) 82.94 77.64 76.18 Less : personal contributions for social insurance 4.21 3.89 4 .00 Residence adjustment -8.25 0.12 -0.02 Net labor and proprietor income (place of residence) 70.48 73.88 72 .16 Dividends, interest, and rent 16.20 14.86 13.80 Transfer payments 13.32 11.26 14.05 Total personal income (place of residence) 100.00 100.00 100.00 Source : Developed from data compiled by the Regional Economics Information System, Bureau of Economic Analysis, U.S. Depart­ment of Commerce, 197 8. Texas generally. Most other forms of economic activity are comparable to state and national averages in their relative importance to the Longview-Marshall area. Since 1930, when the East Texas Oil Field was opened, the extractive industries have been particularly important in the local economy. Kilgore is in the heart of the oil field and was characterized for many years by a large number of oil derricks throughout the town. Relatively few of the picturesque derricks remain today, but Gregg County ranks among the top ten Texas counties in mineral production, with oil and gas accounting for most of the value. In the future lignite is expected to join petroleum and natural gas as a significant contributor to income from extractive in­dustries in the Longview-Marshall area. One lignite mine is in operation in Harrison County, and other potential lignite­mining sites surround the area. Chief Manufacturing Industries The Longview-Marshall area has been a manufacturing center for a number of years. Two of the sixteen largest manufacturing employers in the area were established be­fore World War II: ICI United States Inc. (activated car­bons) and Snider Industries (lumber), both in Marshall. Sev­eral other major industries were established in the area dur­ing and shortly after that war, including two of the largest manufacturing employers in the area-the Thiokol Corp. (munitions) in Karnack and what is now the Marathon Le Tourneau Co. (heavy equipment) in Longview. Manufacturing Plants with More Than 250 Employees Longview-Marshall SMSA, 1978 City Primary Establishment Company products date Karnack Thiokol Corp. Ammunition and rocket motors 1942 Kilgore Kilgore Ceramics Corp. Bathroom fixtures 1948 Nardis of Dallas, Inc. Longview Ladies' apparel 1960 Axelson, Inc. Oil well and gas Continental Can Co., field equipment 1956 U.S.A. Aluminum cans 1970 Marathon Le Tourneau Co. Construction and Pullman Trailmobile mining machinery Truck trailers 1945 1956 Jos. Schlitz Brewing Co. Beer 1965 Jos. Schlitz Brewing Co. Aluminum cans 1974 Stemco Mfg. Co. Motor vehicle Texas Eastman Co. parts Plastics materials 1959 1952 Marshall IC! United States Inc. Activated carbons 1921 Imperial Reading Corp. Dungarees 1946 Snider Industries, Inc. Lumber and wood Scottsville chips 1938 Alcoa Conductor Wire and cable Products Co. conductors 1969 Waskom Fabsteet Co. Steel fabrication 1953 Source: 1977-1978 Directory of Texas Manufacturers (Austin: Bureau of Business Research, 1978). Industrial development activity has continued in Longview-Marshall up to the present. Three of the five largest manufacturing employers in the area came during the 1950s-Texas Eastman Co. (plastics) in 1952, Fabsteel Co. (steel fabrication) in 1953, and Pullman Trailmobile (truck trailers) in 1956. Notable arrivals in the area during the 1960s included the Jos. Schlitz Brewing Co. in 1965 and Alcoa Conductor Products Co. in 1969. Two of the sixteen largest plants in the area have been established since 1970, and both produce aluminum cans. There are over 260 manufacturing plants in the Longview-Marshall area. Only six Texas metropolitan areas, all much larger, have more manufacturing establishments. Total manufacturing employment in the Longview-Marshall area was 18,080 people at the close of 1978, an 18 percent increase over the level five years earlier. Total nonagricul­tural wage and salary employment in the area increased by about 10 percent during the same five-year period. Population and Income Profile The average resident of the Longview-Marshall area is about three years older than the typical Texan. According to Sales and Marketing Management, the median age of male residents of the Longview-Marshall area is 29.6 years, while the median age of female residents is 32.8 years. The corresponding median ages for Texas residents are 27 .0 years for males and 29.3 years for females. Typical household income levels in the Longview-Marshall area are about 8 percent below the state level, with the median household effective buying income of area residents in 1977 estimated at $13,333 in Longview-Marshall and $14,480 for all Texans. (Effective buying income is the total income of all household members after taxes.) Gregg County residents are somewhat more prosperous than their neighbors in Harrison County. The median household effec­tive buying income in Gregg County is $14,223, almost 23 percent more than the $11,573 median household income in Harrison County. Metropolitan Area Characteristics The Longview-Marshall area can be described as having the following characteristics: Age Profile Longview-Marshall SMSA, Texas, and United States (Percentage of population) Longview-Marshall United Age group SMSA Texas States 0-17 29.3 3 1.4 29.S 18-24 11.9 13.7 13.1 2 5-34 14 .1 15.6 15.3 35-49 16.0 16.0 16.3 SO and over 28.7 2 3 .3 2 5.8 Source: Sales and Marketing Management, Survey of Buying Power Data Service, 1978. • A growth rate lower than the state average, but appar­ently increasing. • A high concentration of economic activity in the man­ufacturing sector. • Continuing strong economic activity in the mining sector. • A low level of direct governmental employment and earnings. • A population older than the state average. • Income levels below the state average. Other Significant Factors The combination of advantages that have made Longview­Marshall an industrial area of a significance belied by overall size should continue to influence the growth of the area. Such natural resources as oil and gas have long played a vital role in the development of the area. Lignite resources are now being developed and will contribute to the area role as an energy center. Water and other resources are relatively abundant in East Texas. The area has good surface transportation connections to major markets in Texas and surrounding states. The Longview-Marshall area offers most advantages of metropolitan living without the congestion and other prob­lems associated with big cities. Area employers complain that workers in some skilled trades are in continual short supply, but it is hard to imagine the area not being attrac­tive to many skilled people in colder or more crowded parts of the country. Some amenities are lacking in Longview-Marshall. Sched­uled airline service is limited to commuter service provided by relatively small aircraft, but frequent service is available to the Dallas-Fort Worth area. The area suffers as an air travel market because of its proximity to Shreveport, but residents do have reasonable access to the Shreveport air­port. Several private colleges and universities are in the area, and Kilgore College (home of the Rangerettes) is one of the state's best known junior colleges ; but there is no publicly supported four-year institution-something that many other Texas metropolitan areas have. Yet these problems do not seem to have seriously impaired the economic development of Longview-Marshall. Household Effective Buying Income* Profile Longview-Marshall SMSA, Texas, and United States (Percentage of households) Income Longview-Marshall United (in dollars) SMSA Texas States 0-7 ,999 31.8 27.3 2 5.4 8,000-9 ,999 6 .6 6 .9 6 .6 10,000-14,999 17.7 17.6 18 .0 15,000-24,999 2 8.2 28.7 30.8 2 5 ,000 and over 15.7 19.S 19 .2 *Household effective buying income is the total income of all house­hold members after taxes. Source: Sales and Marketing Management, Survey of Buying Power Data Service, 1978. Tyler: Promising a Rose Garden Charles P. Zlatkovich Tyler is one of the smaller metropolitan areas in Texas but is also among the areas with the brightest prospects for the future. Tyler has several things going for it-a favorable image as a place of beauty and culture, plentiful water and energy resources, proximity to major markets in Texas and adjoining states, and most of the conveniences and services of a big city without the aggravation. Tyler was a settlement during the days of the Republic of Texas and was incorporated in 1846. Named for John Tyler, tenth president of the United States, the community became a trading center for the surrounding fertile agricul­tural area. Tyler's trade prominence was threatened in the early 1870s by the construction of the Texas and Pacific Railway, which passed by Tyler a few miles to the north. Tyler citizens responded by forming the Tyler Tap Rail­road, which opened from Tyler to Big Sandy on the Texas and Pacific in 1877. The Tyler Tap ultimately became the St. Louis Southwestern Railway, popularly known as the Cotton Belt, a major railroad affiliated with the Southern Pacific. Tyler is just south of Interstate 20, midway be­tween Dallas and Shreveport and fewer than one hundred miles from either city. Tyler is probably best known as the city of roses. Center of the nation's largest rose-growing area, Tyler is reported to produce some 20 million bushes annually-more than 50 percent of the national supply. The Tyler Municipal Rose Garden covers twenty-two acres and contains some 38 ,000 bushes in over 500 varieties. The Tyler Area Chamber of Commerce numbers some nineteen commercial rose sup­pliers as members. Population Growth The Tyler metropolitan area (Smith County) has a long history of steady growth. The population of the area in- Charles P. Zlatkovich is Research Associate, Bureau ofBusiness Re­search. creased by about 16 percent during the 1950s and by about 12 percent during the 1960s. Growth has accelerated since 1970; according to estimates by the U.S. Bureau of the Census, the population of the Tyler metropolitan area increased by 12.2 percent between 1970 and 1976. That growth rate is slightly higher than the 11 .5 percent popula­tion increase reported for the state of Texas as a whole. About 58 percent of the growth of the Tyler metropoli­tan area is attributable to movement into the area, with the other 42 percent a result of natural population increase (the excess of births over deaths). These figures are the reverse of those for the state since 1970: Texas is estimated to have grown 42 percent by net migration and 58 percent by natural increase. Tyler ranks eighteenth among the twenty-five Texas metropolitan areas in population, somewhat smaller than Longview-Marshall and slightly larger than Odessa. Among Texas metropolitan areas, Tyler ranks ninth in rate of growth since 1970. The growth of the Tyler area is con- Nonagricultural Civilian Payroll Percentages Tyler SMSA, Texas, and United States 1977 Annual Average Tyler United Category SMSA Texas States Mining 3.9 3.2 1.0 Construction 4.9 7.0 4.7 Manufacturing 27.4 18.4 23.8 Transportation, comm unication, and public utilities 5.1 6.3 5.6 Trade 23.9 24.6 22.3 Finance, insurance, and real estate 4.2 5.6 5.5 Services 18.1 17.2 18.7 Government 12.5 17.7 18.5 So urces: Data for Tyler SMSA and Texas obtained from Economic Research and Analysis Department, Texas Employment Com­mission; U.S. data obtained from Employment and Earnings (Washington, D.C.: U.S. Department of Labor, Bureau of Labor Statistics, April 1977-March 1978). tinuing. The Texas Department of Water Resources esti­mates that the population of Smith County will reach 117,500 by 1980. Employment Structure and Sources of Income The economic sectors that are notable contributors to area income and employment are manufacturing, mining, and services. Despite the presence of the well-known rose­growing industry, the agricultural sector (which includes commercial rose growers) makes a lower relative contribu­tion to income in the Tyler area than it does to total per­sonal income at the state or national level. The contribution of the agricultural sector to the Tyler metropolitan area is higher than the average of that sector for all Texas metro­politan areas, however. Manufacturing contributes much more to the Tyler econ­omy than to the overall state economy, providing more than 22 percent of all personal income in the area. Manu­facturing also contributes a higher percentage to Tyler area income than it does to national total personal income. The mining sector, dominated by oil and gas production, has been a significant part of the area economy since the opening of the East Texas Oil Field in 1930. Mineral pro­duction in Smith County is not particularly high, but Tyler functions as a service center for mining activities in the sur­rounding region. A concentration of health-care activity contributes to the above-average size of the services sector. Tyler has three Percentage of Personal Income by Major Sources Tyler SMSA, Texas, and United States, 1976 Tyler United Source SMSA Texas States Agriculture 0.82 1.55 1.82 Mining 4.22 3.50 1.11 Construction 3.90 6 .22 4 .37 Manufacturing 22 .34 15 .32 19.74 Transportation, communication , and public utilities 5.32 6.08 5.69 Wholesale and retail trade 14.66 15.42 13 .10 Finance, insurance, and real estate 3.09 3 .95 4.01 Services 13.19 11.92 12 .52 Other industries 0 .21 0.26 0 .27 Total private labor and proprietor income 67.75 64.22 62.65 Federal civilian 1.08 3.20 3.28 Federal military 0.12 2 .17 1.34 State and local 7 .2 3 8.05 8.92 Total government earnings 8.43 13.42 13 .53 Total labor and proprietor income {place of work) 76.18 77 .64 76.18 Less: personal contributions for social insurance 3.79 3.89 4.00 Residence adjustment -2.60 0.12 -0.02 Net labor and proprietor income {place of residence) 69.78 73.88 72.16 Dividends, interest, and rent 18.17 14.86 13.80 Transfer payments 12 .04 11.26 14.05 Total personal income {place of residence) 100.00 100.00 100.00 Source : Developed from data compiled by the Regional Economics Information System, Bureau of Economic Analysis, U.S. Depart­ment of Commerce, 1978. large private hospitals and the University of Texas Health Center at Tyler. Partly because of the importance of manufacturing in the Tyler economy, the relative contribution of earnings from sources other than mining and services is lower than the state average, but all sectors are well represented in the area economy. Property income (which includes dividends, interest, and rent) is more important as a source of income in Tyler than it is for the state on the average, in part be­cause of income from oil-producing properties. Transfer payments, which include retirement benefits and welfare payments, are slightly above the state average, but well be­low the national average. Chief Manufacturing Industries A high percentage of the more than 12,000 people em­ployed in manufacturing in Tyler work at the larger area industries. There are about 140 manufacturing plants in the metropolitan area. The 3 largest plants employ more than 1,000 people each. One other plant employs more than 500 people, and 6 additional plants have more than 250 employees each. The three largest manufacturing employers in the Tyler area include a home-grown industry and two major national corporations. The home-grown industry, and the largest manufacturing employer in the area, is Tyler Pipe Indus­tries, which was established in 1935 . Tyler Pipe Industries makes both plastic and cast-iron pipe and fittings and a vari­ety of related products. Two other large manufacturing em­ployers in the area are General Electric, which makes air conditioning and heating equipment in a plant established in 1956, and Kelly-Springfield Tire, a subsidiary of Good­year Tire and Rubber Co. Kelly-Springfield makes tires at a plant established west of Tyler in 1962. A second plant manufacturing air-conditioning equip­ment, operated by Carrier Air Conditioning Co., is the Manufacturing Plants with More Than 250 Employees Tyler SMSA, 1978 Name of Primary Establishment company products ' date Carrier Air Conditioning Co. Air conditioning equipment 1946 R.L. Davis Mfg. Co. Women's apparel 1953 General Electric Co. Air conditioning and heating equipment 1956 Imperial American Co. Furniture ; sport­ ing goods 1962 Kelly-Springfield Tire Co. Automobile tires 1962 La Gloria Oil & Gas Co. Petroleum products 1931 Levi Strauss & Co. Ladies' jeans 1969 National Home Corp. Prefabricated houses 1955 Tyler Pipe Industries Pipes and pipe fittings 1935 U.S.!. Film Products Plastic bags 1961 Source: 1977-1978 Directory of Texas Manufacturers {Austin: Bu­reau of Business Research, 197 8). fourth largest manufacturing employer in the area. The Carrier plant began operations in 1946. Among the other six of the ten largest manufacturing employers in the area, the oldest is La Gloria Oil and Gas Co., a refinery estab­lished in 1931 , shortly after the start of the East Texas oil boom. La Gloria Oil and Gas is a subsidiary of Texas East­ern Transmission Corp. The R. L. Davis Manufacturing Co., makers of women's apparel, began operations in Tyler in 1953. The National Homes Corp. plant, which builds pre­fabricated houses, opened in 1955. Other large industries that opened in Tyler during the 1960s include the Imperial American Co. (furniture and sporting goods), Levi Strauss (women's jeans), and U.S.I. Film Products (heavy-duty plastic shipping sacks and other plastic packaging). Population and Income Profile The average resident of the Tyler area is somewhat older than the average Texan. According to Sales and Marketing Management, the median age of male residents of the area is 29.7 years, while the median age of females is 32.2 years. These medians are about three years above the state levels of 27.0 years for males and 29.3 for females. Income levels in the Tyler area are about equal to the state average. Sales and Marketing Management estimates that for 1977 the median household effective buying in­come, or total income of all household members after taxes, was about $13,840 in the metropolitan area and $14,480 for the state of Texas. Although the median house­hold income is slightly lower in Tyler than it is in Texas, the typical Tyler household consists of fewer persons, so that the median effective buying income per person is actu­ally higher in the Tyler area than in Texas. Total per capita personal income estimates by the U.S. Bureau of Economic Analysis indicate that the per capita income level in the Tyler metropolitan area was about 2 percent higher than the state average in 1976. Metropolitan Area Characteristics The Tyler metropolitan area is characterized by: • A growth rate higher than the state average with above­average levels of net inmigration. Age Profile Tyler SMSA, Texas, and United States (Percentage of population) Tyler United Age group SMSA Texas States 0-17 29.8 31.4 29.S 18-24 11.6 13 .7 13.1 25-34 14.4 15.6 15 .3 35-49 16.2 16.0 16.3 SO and over 28.0 23.3 25 .8 Source : Sales and Marketing Management, Survey of Buying Power Data Service, 1978. • An economy characterized by above-average levels of income and employment in the manufacturing, mining, and services sectors. • Manufacturing employment concentrated in a few well-established industries. • An older-than-average population. • Levels of income and buying power that are higher than the state average. Other Significant Facton Tyler is particularly fortunate among Texas metropoli­tan areas. It probably has the most favorable public image of any Texas city of its size. A favorable public image is an intangible, but nevertheless important, quality for a city to cultivate. Thoughts of other Texas cities of comparable size may bring to mind extremes of heat, dryness, isolation, or poverty; but when people think of Tyler they think of roses. A favorable image is helpful to an area but is seldom suf­ficient to further its development. There are a lot of places in Texas and elsewhere that might be thought of as nice places to live but that offer few ways to earn a living. Tyler has a stable economic base and the resources to support further growth and development without the loss of the characteristics that make the area desirable. Several factors favor the continued growth and develop­ment of the Tyler area. The current size of the area is one. Since 1970, U.S. metropolitan areas with populations of be­tween 100,000 and 250,000 have grown more rapidly than those in any other size category. This growth contrasts to that of earlier years when larger metropolitan areas grew much more rapidly. Americans increasingly believe that the smaller metropolitan areas offer most of the opportunities and conveniences of the larger areas with only a fraction of the problems. Tyler also benefits from its location along a major trans­continental highway (Interstate 20), from the proximity of the Dallas-Fort Worth area, from a favorable climate, and from relatively plentiful water. The combination of avail­able basic resources and a favorable image should combine to make Tyler one of the most promising areas in Texas in the years to come. Household Effective Buying Income* Profile Tyler SMSA, Texas, and United States (Percentage of households) Income Tyler United (in dollars) SMSA Texas States 0-7,999 28.9 27 .3 25.4 8,000-9,99 9 7 .1 6 .9 6.6 10,000-14,999 18.3 17 .6 18.0 15,000-24,999 28.8 28.7 30.8 2 S ,000 an d over 16.9 19 .S 19.2 *Household effective buying income is the total income of all house­hold members after taxes. Source: Sales and Marketing Management, Survey of Buying Power Data Service, 1978. Barometers of Texas Business (All figures are for Texas unless otherwise indicated.) All graphs except the one for nonagricultural employment are adjusted for seasonal variation. Data were compiled from the following sources: U.S. Department of Labor, Texas Employment Commission, Texas Railroad Commission, and Federal Reserve Bank. Data for consumer prices in Dallas and Houston and for oil refining are current through February 1979; all other data are current through March 1979. Consumer Prices (Index 1967=100) 250-----------------------. f Dallas 200 150 100-j,____,_19_7_1-.--~1-9_7_3...---~1-9_7_5...---r1-9_7_7...--...r1-9-79~ Oil Production and Refining (Index 1967=100) 180-....---------------------, 160 140 120 Crude oil producti~~\,., • '" '".._ 80 -j.....-~-19_7_1...---~1-9-73-r---.-19_7_5...---~1-9-77.....--~1-9-7-19 Percentage Unemployed 10.0-..., ....-------------------­ ..• o I " • ' I''' 'i.,, • I .' I I .' ,•''' " ' ' , . 7.5 : ''.I ···~ ~~ United States ; ... I . . I . I .... ~ ~"..._ 5.0 2·5 --l--....1-9_7_1~-r1_9_7_3~-~1-9-7~5...---.-19_7_7-r---.-19-7--19 Industrial Activity (Index 1967=100) 250--.---------------------. 200 150 Total industrial production 100-j,__-,-19_7_1-r----.-19_7_3-r--T1_9_7_5r--r1-9_7_7...--r1-9-7~9 BUREAU OF BUSINESS RESEARCH SECOND-CLASS POSTAGE PAID AT AUSTIN, TEXAS THE UNIVERSITY OF TEXAS AT AUSTIN AUSTIN, TEXAS 78712 5,500-..-------------------------­Total (5,410) Nonagricultural Employment in (In thousands of employees) 5,250 5,000 4,750 4,500 4,250 4,000 3,750 3,500 Trade ( 1,312) Manufacturing (985) Government (971) Service (914) Construction (388)Transportation (346) Finance (305) Mining (190)