CAB - The Economics of Maintaining Land in Agriculture in Fresno County

- Research Publications -

The Economics of Maintaining Land in Agriculture in Fresno County
by
Dennis L. Nef

CATI Publication #960802
© Copyright August 1996, all rights reserved

INTRODUCTION

California's population grew rapidly during the 1980s when the number of people in the state swelled from about 23 million to more than 29 million. While the increased production and consumption activities of these new residents added to the economy, the addition of more than six million people to the state did not come without costs. One cost is the impact of population growth on the amount of farmland available. Where are six million people going to live? If new housing units are built on farmland, the decrease in the amount of farmland available could significantly impact the agricultural and food industry. Thus, a relevant policy issue is the extent to which farmland should be "protected." This report looks at the economics associated with farmland conversion/protection in Fresno County, California.

Fresno County is located near the center of the San Joaquin Valley in one of the richest agricultural regions in the world. The San Joaquin Valley accounts for nearly seven percent of the total value of annual agricultural production in the United States - some $11 billion.

In 1995, Fresno County's 7,500 farmers produced more than $3 billion worth of agricultural commodities on approximately a million accres of cropland. That is a higher value of production than 25 states. Agriculture is often referred to as the major industry in the county and is reputed to generate $3.50 in benefits to the county economy for every dollar in farm sales (Insalaco). Yet it is in this agricultural cornucopia that some of the most rapid population growth in the state is occuring. While the state population grew 26 percent during the 1980s, population in the valley grew 36 percent - an average of 3.1 percent annually. Growth in the San Joaquin Valley is expected to continue at rates 50 percent greater than the state average (see Table 1).

TABLE 1. FIVE YEAR POPULATION GROWTH RATES BY COUNTY ACTUAL AND PROJECTED (PERCENT)
COUNTY 1980-85 1985-90 1990-95 1995-2000 2000-2005 ============================================================== Fresno 13.65 16.96 14.08 11.28 10.19 Kern 20.00 14.29 16.06 12.72 10.45 Kings 15.01 20.73 10.83 9.42 8.53 Madera 20.96 17.69 18.04 13.30 12.24 Merced 19.55 12.52 15.01 13.63 12.88 Sacramento 14.53 17.63 16.25 13.08 10.92 San Joaquin 20.15 16.10 14.84 11.90 10.39 Stanislaus 14.77 23.35 18.51 12.70 11.13 Tulare 15.40 12.16 13.29 11.62 10.58 Valley 16.34 16.72 15.48 12.35 10.72 CA 10.92 13.70 11.33 8.62 7.53 Source: Department of Finance "Interim Population Projections for California State and Counties. Official State Projections".

Such rapid population growth places pressure on existing land uses as cities expand to accomodate new arrivals. Land on the urban fringe is converted from agricultural or other rural uses to residential and industrial uses. In some rural locations, new towns have been proposed. Such changes may explain in part why California land in farms fell 4.8%, harvested land declined by 12.4 percent and irrigated land in the state was reduced 10.2 percent between the 1982 and the 1987 Census' of Agriculture.

While the census data indicate that some land-use changes are occuring, precise data on farmland conversion is not available. California's Farmland Mapping and Monitoring Program (FMMP) identifies land use shifts every two years but has not yet been implemented for all land in the state. For those areas tracked, urban land increased 81,810 acres between 1984 and 1986. Between 1986 and 1988 it increased 106,900 acres and between 1988 and 1990 it increased another 111,391 acres. Of course, not all of the increase in urban uses comes at the expense of farmland and not all farmland is highly productive. Nevertheless, it has been estimated that more than 44,000 acres of California cropland are converted to urban uses annually.

In the past, conversion of other land to agricultural uses offset the conversion of agricultural land to urban uses. This allowed the agricultural land base in California and the San Joaquin Valley to increase and more recently to remain relatively constant. Now it appears that little additional land can be brought into production.

With a statewide cropland base of nearly 11 million acres, conversion appears to account for a decline of slightly less than half of one percent in the amount of cropland per year. Such a small number does not seem to be terribly important given continued growth in productivity from other sources. However, it is argued that the more productive soils are often the ones being converted. Additionally, since conversion is concentrated it is often identified as a significant problem in particular areas. Moreover, agricultural land often has additional dimensions beyond productive value, such as open space, wildlife habitat and cultural aspects. For these and other reasons, the possibility of losing additional agricultural land has sparked a desire among farmers, policymakers, environmental groups and others in government to keep land in agriculture.

Although economics is likely to play a key role in determining which lands are developed and which will remain in agriculture, little information on the benefits and/or costs of development is available. This report presents results from several case studies exploring the use of economic analysis and incentives associated with agricultural land preservation.

WHAT LAND IS BEING CONVERTED?

The state Farmland Mapping and Monitoring Program (FMMP) classifies land as prime farmland, farmland of statewide importance, farmland of local importance, unique farmland, grazing land, urban and built-up land, and other. According to the FMMP, urban and built-up uses of land in Fresno County increased more than 10 percent (7,704 acres) between 1984 and 1990. Roughly 60 percent of the increase in urban built-up uses (4,518 acres) came from agricultural land. There was a total of 18,382 fewer acres in agricultural categories in 1990 compared to 1984. Interestingly, while there were modest decreases in land categorized as prime or of statewide importance, there were offsetting increases in land categorized as unique or locally important farmlands. The net loss was accounted for by decreases in grazing land.

In order to better understand the economics of agricultural land conversion, we examined land that had changed uses between 1979 and 1993 in and around the cities of Fresno and Clovis in Fresno County. Using assessor parcel maps from the two periods, parcels that had been rezoned from agricultural zonings to single family residential and other zoning categories were identified. Distance to the nearest city limit line was also estimated. In total, 10,049 acres originally in 332 parcels were identified.

As shown in Table 2, nearly 60 percent of the acreage was within a mile of the city limits. About one fourth of the land was four or more miles beyond city limits. Clearly, most of the land being converted from agriculture was near urban areas. There is some evidence of leapfrog development as shown by the relatively large percentage (19) in the 4-5 mile category.

TABLE 2. DISTANCE TO NEAREST CITY LIMIT
Distance to city limit acres percent Within limits 2694 27% 1/2 mile or less 1160 12% Between 1/2 and 1 2034 20% One to two miles 919 9% Two to three miles 463 5% Three to four miles 425 4% Four to five miles 1886 19% Over 5 miles 468 5%

Under the Williamson Act, California counties can enter into contracts with land owners to lower property taxes if the landowner agrees to keep the land in agriculture. Farmers are of two minds about the Williamson Act. There is support for it since it lowers property taxes. However, there is some opposition to it from those farmers hoping to develop their property. Thus, areas far from urban areas are entered into contracts while those close to cities where development is occuring are not. As might be expected with 60 percent of the acreage within a mile of city limits, less than sixteen percent of the acreage was under contract. It appears that at least in the study area, the Williamson Act is not used to keep land that is already in close proximity to urban areas in agriculture.

Land Use
The Department of Water Resources (DWR) 1979 cropping pattern maps were used to identify the type of crops produced on the land in the study. Collapsing the numerous cropping codes used by the DWR into discrete sets (e.g. all deciduous fruits and nuts were placed in one category while the actual maps break them out into 12 different crops. Similar adjustments were made to other crops.), the patterns shown in Table 3 emerge.

TABLE 3. CROPPING PATTERNS
CROP ACREAGE PERCENT Subtropical fruits 337 4% Deciduous fruits and nuts 1293 15% Field crops 227 3% Grain and hay 311 4% Idle 326 4% Native vegetation 4364 52% Pasture 271 3% Semiagricultural 10 0% Urban 124 1% Vineyard 181 2% DRY 15 0% Combination 981 12%

The interesting thing about the data in table 3 is that most of the land (52 percent) was classified as native vegetation in 1979. Some land uses on the east side of the study area would be considered to be natural vegetation in any case. However, land that has been previously cropped but idle for more than three years also falls into this category, and it is likely that much of the land in this category is land that had been previously cropped. Only about 20 percent of the land is used for production of high-value crops (fruits, nuts, vineyard).

The Farmland Mapping and Monitoring Program classifications for this acreage were determined from maps. Because the maps could not be overlain to obtain precise measurements, an estimated range of the amount of land in each of the classifications was identified and these are shown in Table 4.

TABLE 4. DISTRIBUTION OF ACREAGE BY FARMLAND MAPPING CATEGORIES
CATEGORY PERCENT urban and built-up 15-19% grazing 23% other 17-24% unique farmland 3-9% locally important farmland 3-4% farmland of statewide importance 7-10% prime farmland 20-27%

Comparing the data in Tables 3 and 4, it appears that some of the native vegetation shown in Table 3 must indeed have once been in farmland. It is also clear that only one-third to one-half the land is in the unique, locally important, farmland of statewide importance or prime farmland categories. Thus, not all land being built up comes from agricultural land, and at least in the areas next to Fresno and Clovis, most of it is not categorized as prime farmland.

This case study seems to show that at least in eastern Fresno County, land being rezoned from agricultural to single family residential and related zonings is generally not prime farmland nor being used to produce high valued products. It is relatively close to already developed areas and is not protected by Williamson Act contracts. Keeping this land in agriculture is not going to substantially increase agricultural production. If open spaces are desired, then alternative forms of policy instruments and a different rationale for protection might be a better approach. In either case, it would be interesting to know something about the magnitude of the benefits and the costs likely to be associated with agricultural land conversion.

MEASURING BENEFITS AND COSTS

There is widespread agreement that agricultural land provides urban areas with amenities so that keeping land in agriculture near urban areas has benefits beyond its agricultural value. On the cost side, the major costs of development are those associated with lost production. It would be difficult to argue that such costs are significant if there is so much agricultural land that development makes very little difference in output or associated amenities or if offsetting increases in farmland elsewhere are possible. If this is the case, it makes little sense to use scarce resources in an effort to protect agricultural lands. On the other hand, if costs are significant, then efforts to protect agricultural lands may well be justified. An accurate measure of the benefits and costs associated with development of agricultural land could guide policy makers in their efforts.

COSTS OF DEVELOPMENT
When agricultural land is developed, certain costs are incurred. The land is no longer available for producing crops so its agricultural value is lost. At first glance, one would think this cost is relatively easy to quantify. In most locations there is a market in agricultural land that establishes its value. Appraisers use such markets and adjust for soil classes, water availability, buildings and other fixed assets on the land as well as other site specific characteristics. The price obtained from this process is usually accepted as the agricultural value. From the perspective of the individual land owner, this is correct. However, from society's perspective there may be other values associated with agricultural land that are not included in this market adjusted price. There is value attached to open space, green surroundings, and the peace and serenity some associate with farmland. These amenity values are "externalities" in that they are valued outside or external to the market. A complete measurement of development costs must include these.

There are also other costs associated with developing agricultural land. For example, new uses will likely increase demands for social programs, public health and safety, highway construction and maintenance, public works, schools, etc. Some of these already were provided to agricultural land, but with conversion to urban uses there will be an increase in support levels.

Quantifying these costs is much more difficult than estimating lost agricultural value. One reason is that the level of these costs depends on the scale of the development. Often, single developments are not large enough to require entire additional units of inputs (a new sewer treatment facility or widening of a highway for instance). Allocating the proper share of the costs is also difficult. While there may be a shift in providers from private to public sectors, such cost transfers are usually irrelevant from the view of the community as a whole unless there is a value associated with this transfer or if there is a difference in the efficiency of delivery.

It is difficult to place the proper cost on a particular activity. Should the average costs associated with current levels of service be used or is the cost of an incremental unit (marginal cost) the proper approach? For example, average costs now being paid by the community for fire protection are likely to be lower than those required by new development. Higher costs associated with newer equipment and more costly construction of new firehouses are among the reasons. Simply extrapolating present costs will not provide an accurate assessment of these costs. Additionally, if these higher costs will be born by all of the community rather than just the new residents, the analyst must remember to measure the increase in costs to existing populations as well as to those moving into the area.

Another cost problem is identifying which other activities are actually costs associated with development and should therefore be included. Will the costs of government go up as a result of this development? Will there be increased air pollution? Will there be a reduction in property values or sales revenues or jobs elsewhere in the community as a result of the new development?

BENEFITS OF DEVELOPMENT
Offsetting the costs are the expected benefits. Development brings increased economic activity which generates increased sales tax revenues, provides jobs, and may improve the community's quality of life. There should be an increased demand for commercial services as well as an increased ability to pay as incomes rise. Land values increase and property tax receipts rise accordingly. Generally, benefits are even more difficult to quantify than are the costs.

TYPES OF ANALYSIS
Several different types of analyses can be used to measure the economic impacts of developing agricultural land. Benefit-cost analysis attempts to consider all the benefits and the costs associated with a particular course of action through time. Alternative projects can be compared by forming a benefit cost ratio (benefits/costs) and comparing the ratios from each project. Since economic measures of all benefits and costs are required, a large amount of data is needed.

Cost effectiveness analysis is often used to compare projects that will have identical or nearly identical outcomes. With comparable results, only the costs have to be considered so that data needs are considerably reduced. That project having the lowest cost is preferred.

Fiscal impact analysis restricts the examination of benefits and costs to those paid for and received by a governmental unit. Thus, only the financial consequences for a city or county are included and it is a limited benefit-cost analysis.

Cost-of-community-service studies have recently been developed to measure the costs of providing public services to various land uses. Costs of public services are apportioned according to demands generated by land use category. These costs are then compared to the revenues from each land use category. Residential costs typically exceed revenues while commercial/industrial and farm/open land categories generate more than they use.

A CASE STUDY OF BENEFITS AND COSTS

Three different agricultural parcels in the path of expected development in the Fresno-Clovis metropolitan area in Fresno County were selected to more closely examine benefits and costs of development/preservation. Estimated costs in terms of foregone agricultural production associated with the development of each site are presented. A limited fiscal impact analysis is also undertaken for each site.

SITE DESCRIPTION
The three properties chosen for this study lie just outside of city boundaries. One site is partially within the Fresno sphere of influence while the other two are outside of current spheres but under study for inclusion by both Fresno and Clovis. Each is approximately two square miles (1,280 acres) and predominantly agricultural in nature but in the path of approaching residential development. The properties will be referred to as the Central, Sanger, and Clovis sites after the school districts in which they are located. The number of parcels in each site and their distribution by size is shown in Table 5.

TABLE 5. Parcel Size by Study Site
Number of Parcels Size (acres) Central Clovis Sanger 1 2 6 4 2 1 13 2 3 9 24 5 4 1 1 0 5 14 7 3 6-10 6 10 14 11-15 3 4 7 16-20 13 6 5 21-25 4 4 8 26-30 2 2 4 31-35 0 2 2 36-40 2 2 1 41-45 0 1 5 46-50 1 0 1 51-55 0 0 0 More than 55 0 0 6

TABLE 6 summarizes land use as documented in a 1986 Department of Water Resources survey. Assessed values and property tax revenues are shown in Table 7. Using land values for different crops in Fresno County (ASFMRA), low and high agricultural values for each site were developed. These are shown along with averages in Table 8.

TABLE 6. Summary of Land Use by Study Site Acreage in each area by Crop Category
Central Clovis Sanger Decidous Fruits and Nuts 1002 850 296 Field Crops 114 0 4 Grain 0 0 41 Idle 7 95 20 Native vegetation 58 35 0 Pasture 44 84 116 Semiagricultural 8 50 100 Truck and berry crops 0 62 7 Urban 1 27 45 Vineyards 46 77 651 Total 1280 1280 1280

TABLE 7. Assessed Values and Property Taxes by Site
Central Clovis Sanger Total Assessed Value $6,840,806 $11,293,478 $21,108,952 Assessed value per acre high $69,159 $129,681 $137,083 medium $5,412 $22,403 $14,896 low $1,237 $2,078 $1,752 Total Property Tax $61,740 $115,122 $218,270
fig8
The Central site includes 1,280 acres west of Fresno. A rural residential area is just to the north. To the east, several new subdivisions are extending the western edge of the city of Fresno. This property has the highest concentration of fruits and nuts of the three study locations with nearly 80 percent of the acreage devoted to these crops - predominantly figs. About nine percent of the land is devoted to row crops, about four percent is in vineyards, and the rest is in pasture, native vegetation or other uses. It has the least amount of built up land of the three sites and the highest average agricultural land value.

The Clovis site lies north of Clovis and east of Fresno. There has been significant development in the Woodward Park area of Fresno just west of this site in the last 10 years. In the last five years, there has also been considerable development just to the south. In 1993 and 1994, there were nearly 20 different subdivisions being built within two miles. About 80 percent of the acreage is devoted to high-value crops. Peaches, nectarines, almonds and figs predominate. Nearly 10 percent is in pasture or native vegetation and just over seven percent is idle. About six percent is in urban or semiagricultural use - much of it in rural residential development in the northeastern corner of the site.

The Sanger site is located to the southeast of Fresno. It is the most developed of the three areas with 11 percent of the acreage already built up. There are a large number of rural residential parcels. Grapes constitute the major crop (51 percent) with other fruits and nuts accounting for an additional 23 percent of the acreage. Nine percent of the land is in pasture and two percent is idle. While less of the land is in high valued crops than at the Central site, the amount of developed land pushes average land values and property tax revenues above those at Central.

COSTS OF DEVELOPMENT
Agricultural Value. If the sites selected are representative of areas surrounding them, greater costs in terms of lost agricultural production occur with development in the Central site west of Fresno. Assuming land values represent the capitalized value of income from the land, an acre of development in the Central area results in losses of $3,840-$7,900. Using a multiplier of 3.5 yields additional losses of $13,445-$27,670 to the county from lost agricultural sales. Thus, total costs in terms of lost agricultural production are in the range of $17,250 to $35,500 per acre. Using a similar methodology and computing averages for the three areas shows that average costs in Central are $26,430, in Sanger, $24,200, and in Clovis, $23,885. Development in Central results in costs 9% higher than in Sanger and 11% higher than Clovis.

Fiscal Impact Analysis. The costs and returns to government agencies of developing a particular site can be quantified (Burchell and Listoken). The costs of streets, sewer, water and solid wastes systems, parks and recreation, police, fire and government must be considered. Revenues from developers fees, increased property taxes, other taxes and charges are also calculated. Based on community plans for adjacent areas, expected growth patterns were determined for each study site. These growth patterns were then used to estimate the needed infrastructure. Costs and revenues from developer fees were then calculated based on current fee structures in Fresno and Clovis. Summaries of the fiscal impact analyses are contained in Table 8. Additional fiscal impact analysis data are available upon request. As shown in Table 8, development costs exceed fees in all 3 locations. The difference is least in Clovis. While this analysis did not include operating costs and revenues expected over time with development, it is clear that revenues will have to exceed operating costs in order to finance development costs.
fig9

A ROLE FOR GOVERNMENT

Externalities in land markets
The benefits from converting agricultural land to other uses obviously outweigh the costs to developers or they would not continue to develop farmland. The losses in terms of productivity must be fully covered or the farmer would not have sold the land for development. If society finds development of farmland to be a significant loss, it must value the losses more highly than the market, which means that the market is missing something. In such a situation, developers are not paying the full costs of development. In economic terms, an externality is involved. An externality can exist if there is some unpriced aspect of the land such as amenity value. Society may value the open space associated with agricultural land but this is not fully taken into consideration in private market transactions between buyers and sellers when the property is developed. Government can overcome this problem by "internalizing" the externality. That is, the full costs of development must be apparent to those making the decisions.

Assessment of benefits/costs under existing programs
Most of the programs discussed here do not act to clarify the costs of development. Amenity values associated with agricultural lands benefit society as a whole. Thus, society as a whole loses if these amenity values are reduced through development. This means that to some extent, development costs are diffuse. However, the benefits of development accrue to those buying and selling the land and are concentrated. Obviously, they will also bear some of the costs, but additional costs borne by others are not a concern. Private assessment of benefits and costs may lead to development when consideration at the societal level would not. Government attempts to balance costs and benefits for society are often made on political, not economic criteria. The decision is in favor of those exercising political power, and that is usually those who are paying or receiving concentrated costs or benefits. For example, zoning ordinances and spheres of influence are sometimes said to be easily changed to accomodate development whether or not it is economically optimal. While cancellation of the Williamson Act is more difficult, political intervention can speed the process here as well. In such cases, government responds to the will of the few rather than the many because the few who stand to be most affected individually are far more vocal than the many who collectively may lose more but individually lose much less.

Towards Explicit Valuation
The use of conservation easements and/or land trusts allows the market to protect land by "locking up" the development rights. Individuals or groups can negotiate with landowners to purchase the rights to develop a particular parcel. Thus, government acting for society can collect fees from all of those benefiting from amenity values and use the funds to preserve land in agriculture by purchasing the development rights. This type of program does "internalize" the externality by introducing amenity values explicitly into the market and forcing those who wish to develop a parcel to consider these "costs." The problem is that governments have yet to come up with a way of identifying how much each household is willing to pay for the amenities associated with agricultural land and then collecting that amount. Thus, "open space" purchases must be made through funds raised by voluntary contributions or sale of bonds. In an era of tight government budgets, it is not likely that much money will be spent on such activities.

An alternative would be to meld several of the existing programs. Government could be used to identify lands in a particular planning area that society would most like to keep in agriculture. Then, government could create a number of development rights and assign these to existing landowners in the planning area. A market for development rights would be established so that the true costs of development are borne by those making such decisions.

Consider a simple market for development rights example. Assume a 20,000 acre planning area that is currently all agricultural. Suppose it is decided that half can be developed and half should stay agricultural. Government planners may even decide which land is to stay in agriculture. Under current programs, this would be extremely difficult. Farmers owning those parcels that can be developed will be able to sell their land for considerably higher prices in the future. The others have the future value of their property limited. This disparity would lead to political deadlock. Formation of a market for development rights overcomes the "windfall/wipeout" scenario. Government issues 80,000 development rights to current landowners (4 per acre) then stipulates that development of an acre of land requires 8 development rights. Those wishing to develop property will have to find four additional development rights for each acre they plan to develop. In order for all 10,000 acres of developable property to be developed, the development rights on the 10,000 acres remaining in agriculture must be purchased and retired. Those who stay in agriculture benefit from the sale (or appreciation) of their development rights while their land is permanently protected from development. Society as a whole maintains the amount of open space desired, and those responsible for developing the land pay the full costs. Government expenditures are limited.

SUMMARY AND CONCLUSIONS

Given expected growth rates in the central San Joaquin Valley, agricultural land will continue to be converted to other uses. Maintaining land in agriculture can be improved by modifying existing programs. However, permanent protection requires a better understanding of the benefits and costs associated with development. Costs vary by site. Around the city of Fresno, it appears that the lowest costs are incurred by developing to the northeast. However, these costs are difficult to derive and changes in underlying assumptions can alter the results. Government can facilitate measurement by incorporating aspects of zoning and conservation easements into a market for development rights approach. Such an approach would require rethinking of goals and educating the various publics involved.

REFERENCES

Insalaco, Cosmo. "Agricultral Corp and Livestock Report." Fresno County.

Burchell, Robert W., and David Listokin. The Fiscal Impact handbook: Estimating local costs and revenues in land development. New Brunksick, N.J. Center for Urban Policy Research, 1978.

{ page top }

CAB Research Publications

{ CATI , also CAB , CFSNR , CIT , VERC }