Chapter 14:Social Impacts Upon Urban Communities

Chapter 14-
Social Impacts upon Urban Communities

by Lay James Gibson

Abstract

Initially, a simple general model is introduced to describe what might happen should irrigated agriculture be eliminated or greatly reduced in the semiarid West. Consideration of this general model raises a number of additional questions: (1) what are the spatial patterns of potential impacts?; (2) what are the employment implications, i.e., how many workers are involved?; and (3) what are the potential scale and other mitigative effects-e.g., to what extent will geographic scale of analysis or previous experience influence the magnitude of impacts? The first question is addressed by analysis of several maps which show the location and relative importance of irrigation agriculture. Data clearly show that irrigation activity is highly concentrated in a handful of states. Furthermore, within these states irrigation agriculture is often spatially associated with metropolitan areas. The second question is approached through case studies which focus on agricultural employment in three states which have exceptionally high levels of irrigation agriculture. "Worst case" estimates of employment dislocation are offered. Even worst case estimates suggest relatively modest dislocations. Finally, speculation is offered as to the scale effects and other mitigative effects that will likely soften dislocation impacts. Impacted regions have opportunities to adjust to diminished water supplies by adapting new farming practices and technologies. Additionally, we can anticipate that impacts will be spread over space, they will occur gradually through time, and governmental intervention can be expected to assist both rural and urban distressed areas. Given the history of outmigration from agricultural areas, additional declines in agricultural population are likely to be of modest magnitude and of manageable proportion when viewed from both the perspective of rural sending areas and of urban receiving areas.

In the years since 1945, total land in farms in the United States has declined by 10 percent, as has acreage of harvested

---

- 332 -

cropland. The number of acres of irrigated land, on the other hand, has increased by 147 percent. In 1945, there was one acre of irrigated land for every 22 acres of cropland; today the ratio is one to nine. The growth of irrigation agriculture in both relative and absolute terms has been an important topic of concern among resource managers, politicians, and the general public. Water transfer projects, groundwater draw-down, and agriculture-urban/industrial water allocation conflicts have all received widespread attention.

Studies of water for agriculture normally focus on supply or demand within an agricultural region; more rare are studies which deal with secondary effects of intraregional shifts in supply or demand for water. This chapter looks at interregional linkages-the spill-over effects on urban regions that might be associated with reductions in supply of water within agricultural regions. Special emphasis is placed on social/demographic impacts. The paper is organized around three main topics: (1) a general model of intra- and interregional relationships; (2) calibration of the populations involved; and (3) speculation about the full consequences of impacts on the urban sector of limited water for agriculture.

A General Model

A general model for understanding what might be expected if quantities of water for agriculture were greatly reduced is fairly simple. Reduction of supply of water would reduce the amount of land used for agriculture (or at least the intensity of use). This, in turn, would reduce labor requirements (both proprietors and wage earners). Displaced workers might live within the impacted region or they might be "seasonals" who live elsewhere, but in either case they would be without jobs. A few might find employment in their local region's agricultural sector or in other industries in their local areas. Others might, at least in the short term, rely on public assistance in their home region, but most are likely to eventually find themselves in an urban residential environment and in an urban job market. It should be noted that whereas many agricultural workers live in rural settings, many others live in, or in close proximity to, relatively large urban centers, e.g., Fresno and Phoenix. These "urban farm workers" look beyond the city for jobs and in many cases, actually live away from home for much of the year. The

---

- 333 -

important notion at this point, however, is that regardless of place of residence, a large share of those who depend on irrigation agriculture for their livelihood would be forced to look toward urban centers for employment if supply of water available to agriculture were greatly reduced. There would be increased reason for these "irrigation workers" and their families to become integrated into the urban mainstream.

A slightly more detailed version of the simple model starts with reduction in direct agricultural employment. These reductions, in turn, produce secondary employment reductions through the multiplier effect; these secondary reductions are likely to be felt first by those in agricultural service activities, e.g., cotton gin employees, and later by those who provide a whole range of goods and services (both public and private) within the impacted regions. Workers made redundant may take other jobs in the impacted region or they may persist, supported by transfer income. But most are likely to head for urban centers with their families. If those displaced already reside in, or adjacent to, urban centers they will likely be forced to become more a part of the urban scene. We might speculate that for every three agricultural families forced to relocate, there will be one "secondary family" that, eventually, is forced to relocate.

Before speculating about specific urban impacts, it is logical to ask questions about the number of workers involved and their locations. Similarly, questions must be asked about the places where water is used. In this paper, critical locations are defined in terms of irrigation. Obviously, all types of agriculture use water. But on the assumption that large water consumers are spatially associated with irrigation and, because irrigation agriculture is frequently labor intensive, attention is focused on leading areas of irrigation agriculture.

Location of Irrigation Agriculture

The relative and absolute importance of irrigation activity is shown on Tables 14.1-14.3 and Figures 14.1-14.3. As can be seen by inspection of Figure 14.1 and Table 14.1, 50 percent of all cropland harvested in seven states is irrigated; irrigated cropland is a conspicuous part of the total in nine other states. These are the states with a high relative dependence upon water for agriculture-states where reductions in the amount of water available would seriously modify the agricultural landscape. It is interesting to note that whereas western states dominate the list, large relative dependence upon irrigation is not a feature exclusive to the West.

---

- 334 -

Figure 14.1
Irrigated Cropland as a Percent of Total Cropland Harvested

---

- 335 -

Figure 14.2 and Table 14.2 also show relative dependence on irrigation, although the pattern is different from that shown in Table 14.1 and Figure 14.1. The variable described here is a sort of "potential landscape change" variable-it suggests that agricultural landscapes, especially in California and Idaho, would change dramatically were irrigation to be withdrawn. The implications are, perhaps, more aesthetic than economic. It is interesting to speculate about the extent to which irrigation creates rural landscapes that somehow enrich the lives of those who pass through them-especially when the irrigation provides a sort of greenbelt around urban centers. A noteworthy feature of these data, when compared to those previously described, is the fact that the same states appear on both lists but their rank order (and interval scale position) is very different on the two lists.

---

- 336 -

Figure 14.2
Irrigated Land as a Percent of All Land in Farms

---

- 337 -

The third measure of location and importance of irrigation agriculture is provided by Figure 14.3 and Table 14.3. In terms of absolute shares, California and Texas are clearly the U.S. leaders with almost one-third of all irrigated land. These two states, along with Nebraska, Colorado, and Idaho, account for over one-half of the irrigated acreage in the U.S. These states, presumably, would be the hardest hit by massive reductions in the amount of water available for agriculture.

Metropolitan Irrigation Agriculture

The distribution of irrigated land among the states of the United States is clearly not even; the same can be said about the distribution of irrigated acreage within each state. In some leading "irrigation states" a large portion of the irrigated acreage is found in metropolitan areas. Tables 14.4 and 14.5 provide information on what might be called metropolitan irrigation agriculture. Whereas data do not specifically describe irrigated agriculture on the urban fringe (some SMSAs include large, remote areas), they do support the assertion that much irrigation activity, and presumably employment, is already spatially associated with metropolitan systems. Many of those working in metropolitan irrigation agriculture may hold nonurban values and beliefs, but they are likely to be more familiar than their

---

- 338 -

Figure 14.3
Location of Irrigated Land: Percentages of U.S. Total

---

- 339 -

---

- 340 -

rural counterparts with urban values and opportunities. Should irrigation activity decline, it is likely that metropolitan irrigation agriculture workers will find the adjustment process much less stressful than their more rural counterparts.

The six states listed in Table 14.4 together account for 61 percent of the irrigated land in the United States. They are a mixed group in terms of the importance of metropolitan irrigation agriculture, but the figures clearly indicate that metropolitan irrigation is a conspicuous feature in some parts of the country. California is clearly the most noteworthy. This state has 17 percent of all irrigated land in the U.S., and 52 percent of it is in SMSAs. Put another way, 9 percent of all irrigated land in the U.S. is in California's SMSAs. Texas, the nation's second ranking irrigation state, has about one-fifth (19 percent) of its irrigated acreage in its SMSAs; 3 percent of all irrigated land in the U.S. is in a Texas SMSA. Fourth ranked Colorado (Table 14.3) also has about one-fifth of its irrigated acreage in an SMSA.

---

- 341 -

Data for a small group of SMSAs with substantial irrigated acreage are given in Table 14.5. California clearly dominates, but Texas and Colorado make a strong showing. Just three California SMSAs alone hold over 5 percent of the irrigated land in the U.S.

Sources of Impacts and their Magnitudes

Ideally, data would be available to describe the number of workers in irrigated agriculture by county in the U.S.; no such data exist. This is unfortunate inasmuch as such figures are needed if we are to know just how large the potential urban impacts might be. At least some light can be put on the problem by estimates of number of workers subject to dislocation. Such estimates will clearly not yield exact numbers, but they will provide at least a general picture of the order of magnitude and spatial distribution of potential urban impact source areas.

It must be recognized that estimates presented in this paper are "worst case" estimates. Estimates of number of employees dependent on irrigation agriculture are very likely higher than the actual number that might be affected by major declines in irrigation activity. The use of "worst case" estimates is standard practice for many types of planning. The military, for example, when evaluating the impacts of a base closure, will use a worst case scenario to assure that plans cover a wide range of possible outcomes. In this study, the worst case is used partly to protect against understatement and partly for more immediate reasons. Specifically, the data which are utilized when making estimates are simply not detailed enough to allow for exact determination of the extent to which individual workers depend upon irrigation.

Source Areas

Although no data exist to describe the number of workers in irrigated agriculture per se, data do exist to describe agricultural employment in counties dominated by irrigation agriculture. For the purpose of generating employment estimates, it will be assumed that all workers in a county with 80 percent or more of its total cropland in irrigation are employed in irrigation agriculture. Obviously such an assumption will overstate the actual numbers inasmuch as workers for noncropland related uses are not discounted. Nor does such an assumption account for variable labor needs of different crops or for farm management practices. But such an assumption is justified in that it will produce

---

- 342 -

at least macro-level estimates which will be of value in establishing the location of broadly defined source areas.

In arriving at direct impact estimates, data on proprietors, hired workers (both paid and unpaid), and agricultural service workers were gathered for key counties in three leading irrigation agricultural states. Specifically, 11 California counties, 16 Colorado counties, and 16 Idaho counties-all with more than 80 percent of their total cropland in irrigation-were inventoried. Variables were defined as follows:

·Proprietors: one per farm was assumed.

·Hired Workers: This term covers paid family workers. Those working 150 days per year or more were discounted by 0.1 to account for some who do not work a full year. Those working less than 150 days per year were discounted by 0.8 to account for those who (a) work less than 5 months per year, and (b) are double counted because they work for more than one employer.

·Agricultural Service Workers: Workers of this type are reported as "paid" and "unpaid" and by the period worked (less than 150 days or 150 days or more). Workers, without regard to pay status, are discounted using the multipliers provided above.

The net result is a series of estimates of full-time-equivalent workers (FTE) that would be displaced if irrigation agriculture were discontinued.

Case Studies

At this point, we shall retreat to case studies of three states which would suffer relatively great losses if irrigation agriculture were discontinued. Ideally, perhaps, we would have a general model which would allow us to produce reliable estimates of labor loss throughout the country. Unfortunately, data are not available that allow the creation of such a model; place-to-place variations in labor utilization are simply too great. Even generation of state multipliers for case studies is hazardous, but statespecific estimates will at least provide figures of general utility. Case studies of California, Colorado, and Idaho are offered to illustrate what the loss of irrigation might mean to three very different areas.

California would clearly be hard-hit by reduced supply of agricultural water. With 8.6 million acres under irrigation, this

---

- 343 -

state is certainly irrigation oriented. California has some 1.3 million persons working in agriculture (including agriculture services). The state's estimated FTE employment (using the procedures outlined above) is 511,601. Using data on California's 11 principal irrigation counties, it is estimated that there are 0.0344 FTE workers per irrigated acre for a total of 295,968 FTE irrigation workers in all counties. In other words, 58 percent of what might be called "full-impact" agricultural employees depend on irrigation agriculture; these are the workers who would be displaced if water is no longer available to agriculture.

Whereas this number is certainly substantial, it must be remembered that California has a civilian labor force of some 11 million-displaced agriculture workers would produce unemployment increases of about 3 percent if all 296,000 workers entered the job market at the same time. Colorado, without irrigation, would lose 70 percent of its estimated 59,000 FTE agricultural workers-a dramatic percentage loss, but less serious than the California case, because Colorado has a smaller irrigated area (3,458,031 acres) and a smaller "body count" (108,766). With a civilian labor force of about 1.4 million, Colorado (like California) would suffer an increase of unemployment of about 3 percent. This is a significant figure, but probably not a disastrous one.

The situation in Idaho is a bit different inasmuch as Idaho does not have the benefit of a broad industrial base as do California and Colorado. With 3.5 million acres of irrigated land-just a bit more than Colorado's figure-Idaho is clearly a ranking state in irrigation acreage. With 114,151 persons in agriculture, this sector enjoys prominence within the state's overall economy. It is estimated that elimination of irrigation agriculture would see the loss of 38,591 of the state's estimated 55,872 FTE agricultural workers-69 percent. Since Idaho's civilian labor force is only some 425,000, irrigation losses of 38,000 or so would represent a major blow to the economy.

Potential Impacts and Potential Mitigative Effects

Metropolitan areas such as Fresno, Sacramento, and Bakersfield in California's Central Valley, and Phoenix, Arizona, already have large communities of agricultural workers living on their margins. Substandard housing, poor sanitation, and other deficiencies are common in such places. The prospect of these

---

- 344 -

areas and intercity areas growing to several times their current populations because of an influx of displaced farm workers is indeed disturbing. And it is clearly likely that displaced workers would turn to urban areas in large numbers if their source of agricultural income were to disappear. Geographic interaction theory suggests that migrants are most likely to select proximate destinations; metropolitan areas in the impacted regions would be the greatest beneficiaries of rural-to-urban migration, at least in the short term. Such migration would certainly produce a new poverty class in the receiving areas-to a large extent, the migration process would reduce levels of rural poverty by increasing levels of urban poverty. A large group without financial resources would now be given the added burden of adjusting to a new and essentially foreign way of life. Social networks would be broken down in sending areas; new networks would need to be built in receiving areas. Potentials for conflict would be numerous as former agricultural workers move to urban areas and compete for low-cost housing and low-paying/low-skill jobs with an existing group of urban poor-an urban poor with longer tenure and better developed urban survival skills.

Evaluation of potential impacts of displaced agricultural workers on urban areas is tricky for two types of reasons. First are a series of items related to agricultural production practices and technologies. Second are a number of considerations that define the nature of the impact process itself. It must be remembered that the estimates of workers affected are "worst case" estimates of the total number that could potentially be displaced. It is highly unlikely, however, that rapid, wholesale displacements would occur.

Practices and Technologies

All regions, including those now experiencing diminished supplies of water for agriculture, have a number of ready options available to prolong their life as productive agricultural areas. Perhaps the most obvious is a move to (or return to) dry farming. Many areas which now depend heavily on irrigation were originally developed as areas of dry farming. Second, many areas might turn to crop substitution for their salvation, i.e., more water efficient crops might be introduced to replace the heavy water users now cultivated. Third, new field preparation techniques which encourage water conservation, e.g., laser leveling, might be employed. Fourth, new water-efficient irrigation practices might be introduced. Fifth, water transfers or even deep wells might prove to be the answer in some areas.

---

- 345 -

The options selected will depend upon the willingness of an area's farmers to adapt new practices and technologies and on the ability of the natural environment to support them. But at least in many places, agriculture can be expected to persist even in the face of diminished water supply.

The Impact Process

The nature of the impact process itself is also important to consider.

1. Geographic Scale . Even "worst case" estimates of nationwide impacts might suggest a situation that would be serious but not disastrous. However, when these same estimates are applied to specific urban areas, potentials for negative impacts will often be substantial, i.e., impacts are likely to be concentrated in a relatively small number of urban areas in California and the Southwest. California would clearly suffer the greatest impacts. Further, water here is often more than simply an amenity that allows for diversity in crops or increased yields-it is often essential for the very existence of commercial agriculture. Perhaps one bright spot for a place like California is the fact that many of its farm workers are now employed in close proximity to metropolitan areas-they are already somewhat "street wise." A number of other states (see Tables 14.1-14.3 and Figures 14.1-14.3) would also be impacted, but in most cases impacts would be less severe because the number of workers involved is less and because the opportunities for alternative employment in agriculture within the general area are greater.

2. Timing. Obviously, the time period involved is critical. If all irrigation agriculture were eliminated simultaneously, impacts would be dramatic. If, on the other hand, reductions in irrigation activity were spread over a number of years (as they certainly would be), the adjustment process would be much smoother. In fact, one could argue that the impacts of a relatively long-term rural-to-urban relocation would be little more than an extension of shifts that have been in evidence since the turn of the century. Further, it might be argued that because numbers of people are relatively small, especially in relation to the size of receiving centers, and major receiving centers, e.g., western metropolitan areas, are usually high-growth areas anyway, new populations could be accommodated with only minor dislocations.

3. A Further Note on Timing. The history of decline of agriculture population and labor force merits additional comment because that trend serves as a standard against which the

---

- 346 -

severity of irrigation losses can be measured. In the 40 years between 1940 and 1980, U.S. farm population declined by almost 500 percent or 24,500,000 people (Table 14.6). During this period, U.S. population grew by 72 percent to the 1980 figure of 227 million. During the period 1950-1970 alone, farm population declined by over 13,000,000. Every state experienced declines during these years (Figure 14.4); with the exception of Texas, the biggest losers were not states with large areas in irrigation. Nevertheless, the six leading irrigation states did lose over 1.6 million farm population-not a trivial number.

A somewhat different accounting system (Table 14.7) produces another picture of decline that is essentially consistent with the one just presented. Agriculture's role in the U.S. labor force has declined significantly in both absolute and relative terms since the late 1940s. The agricultural labor force has declined by about 4.6 million; agriculture presently directly supports only about 3 percent of our total labor force.

Using figures presented elsewhere in this chapter, we can produce a rough estimate of irrigation agriculture employment for the nation. An estimated figure of 1,000,000 is offered for purposes of discussion. Such an estimate almost certainly overstates the true number, but such overstatement is consistent with the "worst case" approach. But the important point is this-given the history of decline of both "agricultural employment" and "labor force," the worst case still presents us with a situation which is considerably less dramatic than our actual experiences in the years following World War II.

4. Alternative Livelihood Opportunities. Not all displaced workers will leave the impacted region. At least some will shift jobs without leaving their present place of residence. Still others will stay in place and substitute transfer income, e.g., public assistance, for earned income. Finally, some secondary breadwinners may simply drop out of the labor market. Aggregate family income would suffer in such cases, but if the primary breadwinner's job is secure there will be little reason for at least many of them to move one.

5. Diffusion of Impacts. Some workers are migrants with home bases in many parts of the country and even in foreign countries. Their expenditures in areas where they are employed are usually modest at best; their absence should go largely unnoticed in employing regions. Since sending regions are widely dispersed, impacts above the family level should be minimal here, too.

---

- 347 -

---

- 348 -

Figure 14.4
Farm Population Declines Between 1950 and 1970

---

- 349 -

6. Political Considerations. The farm lobby is still strong in the United States. There is every reason to believe that even if political pressure cannot make water, it can make waves-waves which will produce support for projects and programs designed to extend the agricultural life of impacted regions.

In short, both farm operators and farm workers have numerous opportunities for minimizing or at least delaying impacts. The full force of any potential impacts will certainly be blunted by the fact that they will be spread through time and over space. But perhaps the greatest comfort comes from knowing that, after decades of outmigration from agricultural areas, there are just not that many people left that can be displaced.

Discussion:
Darrell K. Adams

During the past decade, the research area subsumed under the title "social impact analysis" (SIA) has increasingly seen a focus in approach which suggests, if not the beginnings of maturity, at least the close of the adolescent period of development. We began this enterprise with a plethora of approaches which had, as a common feature, a kind of encyclopedic examination of the social world, in an effort to assure that nothing of importance escaped scrutiny. Now, the focus is on narrowing the range of variables to those significant to a decision.

A general paradigm for SIA now usually includes some provision for assessing changes in (1) employment, (2) income, (3) population, and (4) the social meaning those changes might have from a variety of perspectives. The first three constitute "impact" analysis (i.e., an analysis of a measurable change directly attributable to some other changed circumstances), while the fourth would be a "social effect" analysis, or an assessment of social meanings and interpretations of impacts.

Gibson has presented a model designed to examine the urban impacts of reduced water availability for agriculture in the West. As he indicates, the model is quite straightforward. A reduced water supply would tend to reduce agricultural production with reduced labor requirements displacing workers which, in turn, would result in residential relocation (rural to urban). This general paradigm is used to organize data so as to answer some important questions dealing with the spatial location of areas of severe impacts in terms of numbers (both absolute and as a proportion of workforce). Thereby, contributions are made to:

---

- 350 -

(1) scoping the potential problem, and (2) spatially locating the potential problem areas. While Gibson does not offer his analysis as the definitive answer, his contribution is nonetheless instructive. The potential for negative impacts of some significance in a relatively small number of urban areas in California and the Southwest is noted. He then comments that even these potential problem areas are subject to a variety of reasonably mitigating conditions.

One might be inclined toward a relatively sanguine view of the urban impacts of a reduction in water supply for irrigated agriculture in the West from Gibson's analysis. However, such a view could be somewhat premature. The model is yet in an early stage of development, and emphasis is placed on employment (number and location of jobs) and population movement impacts. These are but two of the four general classes of variables useful for analyzing social impacts and effects.

As the next stage of development of the model, it would be helpful to trace the income variable through the model. Reduction in crop intensification or reversion to dryland practices on the same amount of land should result in reduced farm income. Reduced farm income, in turn, would have a variety of other impacts. It might, for example, eventually result in reductions in urban support facilities and services. Such reductions would adversely impact both long-time urban residents and newly displaced rural-to-urban migrants needing such support. It is possible, especially during periods of poor economic conditions, that such impacts could be significant and troublesome. The actual significance of this and other multiplier effects would need to be examined in further development of the model.

The other major class of variables-the social effects analysis-would be more difficult to assess. If reasonably solid assessments of the spatial and temporal locations of employment, income, and population impact variables were available, they could be presented to the potentially affected publics in order to obtain their interpretations of the social meanings of such impacts. This analysis becomes more complex because, as we have found in site-specific social analyses, each significantly affected group of the public is likely to have a different interpretation of the social meaning of a particular event. In the same way that a one-dollar change-in-income will have different meanings for a wealthy person than for a poor person, the projected impacts of reduced water supplies for irrigated agriculture upon employment, income, and population will have different social meanings for different groups and communities.

---

- 351 -

It is important to recognize that, regardless of the practical difficulties of completing the social effects analysis, it is only at this stage that a thorough understanding of the social importance of the impacts is attained. Toward this end Gibson's model has made a useful contribution.

Discussion:
Evan Vlachos

Gibson's presentation addresses the major question of trying to anticipate the general impacts of limited water for agriculture in the semiarid West and, more specifically, the potential effects on urban localities. One must really relate this particular chapter to earlier community migration studies, as well as to the vast literature that developed after the National Environmental Policy Act. The latter, in particular, has contributed to elaborate conceptual and methodological frameworks concerning the assessment of direct and indirect impacts of programs, projects, or activities on the surrounding environment. In this context the search for an accounting of all relevant impacts became not only a legal requirement, but also raised important theoretical questions leading, perhaps, to a more cogent interdisciplinary model incorporating the sets of circumstances and web of interactions that contribute to both short-term as well as far-reaching consequences.

Gibson bases his argument on a rather greatly simplified "model" in which, as a result of the reduction of water supply and with attendant reduction of labor requirements, displaced workers (particularly those living around metropolitan areas) will increase the population of surrounding urban localities. Such a "model" must obviously be further elaborated by considering three major subdimensions that eventually could more accurately describe both impacts and long-range consequences, namely:

a) the classical demographic understanding of migratory movements in terms of "push and pull" factors (i.e., reasons for out-migration as well as forces of attraction of specific localities);

b) the overall time as well as the rate of change, i.e., both the entire time horizon (whether temporary or permanent) and the rate of transformation (whether rapid or slow);

---

- 352 -

c) the composition of migratory stream, especially in terms of ethnic, racial, sex, and other components.

In Gibson's presentation the model concentrates on calculation of impacts in agricultural states as a result of the number of workers who might migrate if water supply is reduced in the irrigated West. Quite correctly, Gibson points out that since no relevant data exist, guestimates must substitute for precise information. He points out that, depending on the particular case, although there may be significant numbers of potential migrants, the overall movement to urban localities may not be particularly disastrous.

In trying to supplement Gibson's discussion I would like to point out additional items that could be meaningful.

First of all, one should take into account not only the "formal" agricultural population but also (especially in the case of California) "invisible" workers. Reference should be made to alien workers who have not traditionally been counted and whose contribution to the surrounding economies may be quite significant (directly and indirectly).

The distinctions made by Gibson between Colorado and California, on the one hand, and the expected impacts in the case of Idaho, on the other hand, depend on the economic and sectorial composition of these three states. Since Idaho does not have the broad industrial base of California and Colorado, it should be much more significantly affected (although again the number of alien workers may alter the extent of expected effects). Yet, despite their historical backdrop, many states in the West have been recently characterized by more diversified economies and more resilient localities, especially urban centers of high absorptive capacity (such as the emerging megalopolis of Colorado's Front Range).

In addition, throughout Gibson's chapter there seems to be some vacillation as to the ultimate consequences of a reduction of water for agriculture. While in certain parts it is emphasized that there may be some significant consequences, elsewhere (see notably the conclusion) it is pointed out that the potential urban migration is simply part of the continuous urbanization and suburbanization of American society and of diminishing farm employment. Heavy automation in American agriculture and the emergence of an efficient agribusiness industry may also account for hypothesized minimal impacts, certainly not of the proportions of previous migratory movements such as those felt during

---

- 353 -

the drought of the 1930s and the exodus associated with the Depression era.

All in all, Gibson has correctly pointed out both the inadequacies of the present data as well as the essential parameters of a broad model that could account for the relationship between existing population concentrations in agriculture and the increasing urbanization in the West. If one is to examine the importance of irrigated agriculture and the potential of far-reaching social consequences stemming from water reductions, further distinctions and elaborations must be made with regard to the type of affected populations, particularly in terms of the demographic characteristics of such states as Colorado and California; in terms of the ability of cities and of surrounding urban and semiurban localities to absorb the limited number of workers currently employed in irrigated agriculture; and in terms of larger social policies that can cushion the effects or could mitigate earlier disastrous voluntary and nonvoluntary population movements.

The above remarks should not be construed as implying that there can be no negative impacts from a potential reduction of water in agriculture. Indeed, what is most important is not the total number of people who may be affected. More important are the far-reaching social consequences-and the transition from a predominant ideology and culture that still emphasizes a balance between rural hinterland and urban localities, to one of a highly urbanized and intense postindustrial economic base. What needs to be recognized vis-a-vis the urban impacts is to what extent urbanites, especially refugees from the humid East, are capable of understanding the cultural heritage of irrigated agriculture in the West. Can they respond with sensitivity to the need for coexistence with irrigated agriculture-that agriculture being both a means of survival and a way of life that has characterized the salubrious environment of the western United States? Otherwise, an alternative view (perhaps even a future scenario) may be a totally transformed "Sunbelt" characterized by cybernetic industries, hydroponic farms, water for energy development, and a playground for the rest of the nation, with only dim memories of irrigated agriculture.