Protecting Rural Amenities Through Farmland Preservation Programs

16 Pages·2003·1.33 MB·English

The World’s Largest Open Access Agricultural & Applied Economics Digital Library

This document is discoverable and free to researchers across the
globe due to the work of AgEcon Search.

Help ensure our sustainability.

Give to AgE con Search

AgEcon Search
http://ageconsearch.umn.edu
[email protected]

Papers downloaded from AgEcon Search may be used for non-commercial purposes and personal study only.
No other use, including posting to another Internet site, is permitted without permission from the copyright
owner (not AgEcon Search), or as allowed under the provisions of Fair Use, U.S. Copyright Act, Title 17 U.S.C.
Protecting Rural Amenities Through
Farmland Preservation Programs
Cynthia J. Nickerson and Daniel Hellerstein
We investigate what farmland preservation programs reveal about the importance of protecting
different rural amenities. An extensive content analysis of the enabling legislation of various farmland
protection programs suggests wide variation exists in the protection of amenities. An analysis of 27
individual Purchase of Development Rights (PDR) programs’ selection criteria suggests these pro-
grams favor preserving amenities that are jointly provided by cropland and livestock operations. These
PDR selection criteria also reveal unique preferences regarding the spatial patterns of preserved
agricultural lands. Variation in relative weights given to protecting most parcel characteristics in PDR
programs is not easily explained by factors that characterize areas experiencing farmland losses.
Key Words: farmland preservation, rural amenities
Despite the relatively small fraction of the American agricultural production, such as the Conservation
landscape dedicated to urban uses, there is growing Reserve Program, suggests reasons not directly
concern about the disappearance of farmland in related to agricultural production may also be impor-
some parts of the country. This concern is reflected tant—such as the desire to maintain “rural amen-
in the adoption of an expanding array of farmland ities” associated with agricultural land uses (e.g.,
protection programs by nonprofit organizations and scenic views, wildlife habitat, agrarian cultural
by county, state, and federal governments. Individ- heritage, and open space). Although the effective-
ual states now spend millions of dollars annually to ness of farmland protection programs in addressing
protect farmland, through such mechanisms as use sprawl prevention concerns is beginning to receive
value assessments and Purchase of Development attention (Heimlich and Anderson, 2001; Nicker-
Rights (PDR) programs. The 2002 Farm Bill pro- son, 2001), few studies have provided a compre-
vides evidence of growing concerns at the federal hensive assessment of what rural amenities are
level. Funding for the federal farmland protection likely to be protected through farmland protection
program increased more than ten-fold, from about programs.
$53 million total for the six years 1996S2001 to In this paper, we seek to increase our under-
nearly $600 million total over the six years begin- standing of the rural amenities protected through
ning in 2002. farmland protection programs, and how the provi-
While interest in protecting farmland may arise sion of these amenities varies across the country.
in part from desires to maintain agricultural activ- We do this by examining farmland protection pro-
ity, the existence of federal programs to limit gram legislation, and the methods used by one type
of program to prioritize parcels for preservation.
Underlying this research is the desire to understand
Cynthia J. Nickerson and Daniel Hellerstein are economists in the what positive benefits, including the provision of
Resource Economics Division, USDA-Economic Research Service. The
authors are grateful to Charles Barnard for his invaluable assistance, and rural amenities, are sought by and provided to the
to William Anderson, Joe Dewbre, and Robert Johnston for helpful public through farmland protection programs. How-
comments. This paper extends an ERS project which seeks to characterize
ever, as government programs could very well be
public preferences for the preservation of rural amenities. The views
expressed here are those of the authors, and may not be attributed to ERS. influenced by interests other than the general
This paper was presented at the Land Use Policy Workshop of the public’s preferences (e.g., farmer interest groups’
Northeastern Agricultural and Resource Economics Association annual
meetings, Harrisburg (Camp Hill), PA, June 9S11, 2002. preferences), examination of farmland protection
Agricultural and Resource Economics Review 32/1 (April 2003): 129S144
Copyright 2003 Northeastern Agricultural and Resource Economics Association
130 April 2003 Agricultural and Resource Economics Review
programs is unlikely to yield precise measures of context-specific circumstances affect survey out-
the public preferences for different rural ameni- comes. For example, in any given area, the number
ties. and types of programs that exist to protect rural
Most of the existing research on farmland pro- lands are numerous and complex, and may have
tection programs and amenities has focused on overlapping objectives. Respondents would have to
people’s willingness to pay to protect varying be remarkably well informed about the plethora of
amounts of farmland using contingent valuation programs and their competing or overlapping goals
methods (Bergstrom, Dillman, and Stoll, 1985; to be able to express preferences over goals of any
Beasley, Workman, and Williams, 1986; Halstead, single rural land protection program in a survey.
1984; Bowker and Didychuk, 1994; Rosenberger Also, the extent of land remaining in agriculture
and Walsh, 1997), or their willingness to pay to live at the time of the survey can be expected to influ-
on or near protected farmland as revealed via ence stated preferences about farmland preservation
hedonic approaches (Irwin and Bockstael, 2001). goals. For example, only 7% of the land in Rhode
While these studies consistently conclude that people Island was in a farming use when that survey was
are willing to pay to protect farmland (and do so, as conducted (Kline and Wichelns, 1996). Agriculture
observed through increased property prices of adja- was a predominant land use in the Chicago collar
cent houses), much less is understood about exactly county study area (Krieger, 1999). In the North Car-
what objectives people seek when they support olina study, one-third of the county was urbanized
farmland protection programs. (Furuseth, 1987). Further, the presence of other
One study that did examine public preferences institutions affecting rural land preservation, as well
regarding the goals of farmland preservation was as a limited understanding of development patterns
conducted by Kline and Wichelns (1996). In a and rates, could contribute to difficulties research-
survey of Rhode Island residents, they found more ers report survey respondents experience when
support for protecting groundwater, wildlife habitat, asked to trade off which farmland attributes ought
and natural places than agricultural objectives such to be protected specifically through farmland pre-
as protecting local food supplies or maintaining a servation programs.
farming heritage. In an extended study using the Our goal is to investigate what the farmland
same data, Kline and Wichelns (1998) also found protection programs reveal about the importance of
that on average, Rhode Island residents preferred rural amenities. In doing so, we take a broad view,
preserving fruit and vegetable farmland with public and consider programs in place across the country.
access, woodland without public access, and fruit We adopt an approach based in the economics of
and vegetable farmland without public access—but public choice which describes how public prefer-
these preferences varied depending on whether ences influence government program design.
respondents had environmentally oriented attitudes, First, decisions of idealized farmland preservation
preferences for maintaining rural character and open agencies are modeled, demonstrating how public
space, or farming-oriented attitudes. preferences influence the decision process of which
Survey results in one county in North Carolina amenities to protect, and also how the presence of
revealed support for farmland preservation pro- other rural land protection programs affects
grams arose as much from desires to protect food decisions. We then model the decisions made by
supplies and farming heritage as from a desire to agencies administering a particular type of vol-
protect the environment by keeping land in open untary farmland preservation program, namely a
space (Furuseth, 1987). In Chicago collar counties, Purchase of Development Rights program (also
the most important reasons focus groups identified known as Purchase of Agricultural Conservation
for protecting farmland were ensuring future food Easement program). This model demonstrates how
supplies, protecting family farms, and controlling the agency’s method for choosing which parcels to
development (Krieger, 1999). In a Delaware sur- preserve in a voluntary program can reveal infor-
vey, the most important reasons for protecting land mation on the relative importance of protecting
included maintaining the agricultural way of life, different parcel characteristics, which, in turn,
access to locally grown products, and protecting determines the types of amenities most likely to be
water quality (Duke and Hyde, 2002). protected.
In addition to being restricted in scope (i.e., only The analysis of data involves two approaches.
residents from one county or state are surveyed), Using a content analysis approach, enabling legis-
these studies provide limited information about how lation is examined for a wide variety of farmland
Nickerson and Hellerstein Protecting Rural Amenities Through Farmland Preservation Programs 131
protection programs in the lower 48 states. This could capture the value of rural amenities (i.e., if
analysis reveals the extent of spatial variation in they could require consumer payment), the profit-
objectives, and the array of rural amenities that ability of maintaining their land in a farming use
states seek to protect through these programs. Then, would be increased. This would tend to postpone
using data on 27 individual PDR programs, we conversion of the farmland to nonfarm uses, since
investigate variation in the relative weights placed alternative uses would have to generate higher
on various farmland attributes. Although the analy- income streams in order to bid the land away from
sis does not lend itself to definitive conclusions on a farming use.
the values individuals have for different amenities, Alternatively, even if rural landowners could sell
we are able to offer some general findings on the rural amenities to individual consumers, to the
importance of the various types of amenities most extent rural amenities are nonrival in consumption,
likely to be preserved in farmland preservation pro- they will be underprovided. Individuals purchasing
grams. a rural amenity will only consider their own well-
being when assessing tradeoffs between this rural
amenity and other commodities, even though their
Background: Rural Amenities, Public
purchase of a rural amenity would provide a good
Goods, and Farmland Protection
which can be enjoyed by all. Consequently, each
individual would choose to purchase too small an
Broadly defined, rural amenities encompass a vari-
amount of the rural amenity relative to the social
ety of goods and services that provide utility to
optimum (Samuelson, 1954).
consumers, and whose provision requires a rural
Several mechanisms have evolved to help correct
setting. Rural settings refer to lands that begin at the
the market failure that arises when rural lands are
edges of cities, where the landscape is not domin-
ated by human settlement.1 Agriculture and forest- developed too soon. These include private (non-
governmental organization) initiatives such as
land are examples of predominant land uses one
private rural land trusts, formed to accept charit-
would find in a “rural setting.”
able donations of rural lands—the entire interest
What distinguishes the goods and services we
in the land, or a partial interest such as the devel-
call “rural amenities” is that a rural landscape is
opment rights associated with the land. These
involved in their production. Generic commodities
mechanisms also include initiatives where a set
like bushels of corn purchased from the market are
of individuals join together to collectively pro-
not considered to have “rural amenity” value, even
tect farmland. Under some circumstances, this
though they are likely to have been produced in a
latter strategy can result in the efficient provision of
rural setting—because the value of such commodi-
rural amenities. For example, when amenities are
ties is derived from attributes such as the nutritional
highly localized, the flow of benefits could be
content and flavor of the corn, and not specifically
retained by a limited group. An illustration would
from where it was produced.
be the decision to subdivide a farmed parcel and
Many rural amenities have the distinction of not
situate house lots in a circular pattern around a core
being readily tradable in markets. A number of
of open land containing the original farm buildings,
these goods and services, such as scenic farm views
which continues to be farmed. However, to the
or the agrarian cultural heritage of an area, have a
extent the rural amenities provided by the protected
public goods nature and are distinguished by non-
farm yield benefits (such as scenic views) to
excludability and nonrivalness in consumption.
noncontributors, the problem of underprovision is
The potential underprovision of rural amenities
likely to remain.
from agricultural land arises because of their public
The most common approach to the protection of
goods nature. When rural amenities are nonexclud-
farmland is through government programs. In the
able, the owners of rural lands will be unable to
next section, our models of the decisions of farm-
profit from their production, because consumers are
land preservation agencies are introduced. We dem-
not required to compensate the producer for their
onstrate the role played by public preferences and
consumption of the amenity. If rural landowners
illustrate why the focus of farmland protection
programs is location-specific. Our discussion high-
1 It is useful, especially in the United States, to add another boundary— lights the implications of the decisions for the
where the wildlands begin. Since “city’s edge” and “start of the wild- protection of rural amenities through farmland
lands” are rarely sharply delineated, this is meant to be suggestive of
landscapes defined, but not dominated, by a human presence. preservation programs.
132 April 2003 Agricultural and Resource Economics Review
Conceptual Models of Rural Amenity preservation, considering other agencies’ actions is
Provision via Farmland Protection particularly important because of the existence of a
Programs plethora of other rural land preservation and con-
servation programs which also provide amenities.4
The analysis of government decisions is closely We also proceed under the premise that these pro-
related to the economics of public choice, which grams are politically palatable to farmland owners,
postulates that voters support programs which max- and assume these interest groups’ preferences are
imize their utility (Stevens, 1993). When choosing not over-represented relative to the preferences of
public goods, individuals partially act as if they the general public in agency decisions. This
were making choices affecting their own consump- approach allows us to focus attention on the effects
tion of goods and services (Reichelderfer and of individual preferences for amenities on program
Kramer, 1993), and partially as citizens expressing design.
their values (Margolis, 1992; Quiggin, 1997). In In making optimal decisions regarding farmland
either case, preferences are being acted upon. protection, the agencies can be expected to consider
Moreover, recent research on environmental the utility individuals derive from rural amenities.
issues shows bureaucracies are responsive to the Using an indirect utility function, individual utility
desires of their constituencies, and also that public can be modeled as a function of income, the price
preferences shape program design (Cropper et al., of “use-value” rural amenities, and the quantity
1992; Yates and Stroup, 2000; Hewitt and Brown, of “existence-value” rural amenities. The price of
2000; Weingast and Moran, 1983). Hence, although “use-value” rural amenities, P, conditions the con-
i
program design and adoption may be influenced by sumption of goods and services that individual i can
political groups representative of farmland owners obtain through the purchase of indirect inputs. In
(since the programs directly affect this interest particular, these include rural amenities such as
group), it is equally likely these programs are corre- “outdoor recreation” and “sightseeing,” which the
lated with popular preferences.2 individual can travel to obtain (the cost of travel
being the indirect input). Hence, P are individual
i
Farmland Protection Agency Decisions specific, and depend on where the individual lives
in Theory relative to rural landscapes that produce these kinds
of rural amenities. “Existence value” rural ameni-
When farmland protection agency decisions con- ties, Z, are “pure” public goods. These include
sider the utility individuals derive from rural amen- “maintaining biodiversity in the local ecosystem”
ities, as well as the presence of other protection and “knowing that farming as a way of life con-
programs that provide amenities, the decisions tinues in one’s community.” Since these amenities
reveal several interesting findings regarding the are both nonrival and nonexcludable, their quantity
types and values of amenities likely to be provided is the same for all individuals.
in farmland preservation programs. Although food Indirect utility can be defined as:
security, urban planning, and rural development
gwohaalts faorlelo awlsso w oeb sjeetc tthiveesse oofth eexr icsotinncge rpnrso garsaidmes a, nind (1a) Vi(yi,Pi,Z)'jδtE[V(yit;Pit,Zt)],
t'0
focus on the provision of rural amenities as the goal
where δtE[V(·)] = the discounted expected indirect
of the agency’s farmland protection program.3
utility of individual i, and the expectation is with
Our stylized model starts with the notion that
respect to future values of y, P, and Z; y = a
farmland protection occurs within a broad set of it
sequence (over t) of personal income values; P = a
rural land conservation programs (Rosenberger, i
sequence (over t) of the price vector for “use-value”
1998). In the case of rural amenities and farmland
rural amenities; and Z = a sequence (over t) of the
t
quantity vector of “existence-value” rural ameni-
2 Although principal-agent factors can limit the efficiency of any gov- ties. Prices and income are normalized by the price
ernment program (Niskanen, 1991), in practice, government provision of
of a numeraire good.
public goods is common, and may often be nearly optimal (Migue and
Belanger, 1974).
3 Clearly, food security, urban planning (i.e., sprawl prevention), and
rural development are often important goals of farmland protection pro-
grams. However, it can be argued (given alternative mechanisms exist to 4 A theoretical appendix that discusses optimal decision making, when
provide these benefits), that the provision of rural amenities may be the several land conservation programs are active, is available from the
most compelling reason for farmland protection (Hellerstein et al., 2002). authors upon request.
Nickerson and Hellerstein Protecting Rural Amenities Through Farmland Preservation Programs 133
Farmland protection agencies will consider the both the spatial distributions of rural amenities and
value people place on rural amenities. The farmland residence patterns. That said, a perusal of the model
protection agency seeks to maximize over all i indi- helps to highlight a number of points.
viduals:5 First, equations (1a)S(1e) formalize the notion
that there is no such thing as a single “rural amen-
(1b) jWTPi, ity.” Rather, there are a number of rural amenities,
i with different types of rural lands providing them in
where different quantities. For example, open space and
wildlife habitat can be provided by any rural lands,
V(y &WTP,y˜ ;P1,Z1)'V(y,P,Z). while pastoral scenic beauty and the cultural heri-
i0 i i i i i
tage associated with farming as a way of life are
Here, WTP is individual i’s willingness to pay to uniquely provided by farmland.7 In particular, we
i
protect rural amenities, y is the current year in- can identify a subset of “existence-value” rural
i0
come, y˜ 'y ,y ,...is the vector of future income, amenities which are uniquely provided by farmland
i i1 i2
and P1andZ1(the sequence of P and Z, after pro- as Zf, and the prices of “use-value amenities” which
i i
tecting farmland) are determined by: are uniquely provided by farmland as Pf. Similarly,
we define amenities (and prices) that may be pro-
(1c) P1'G (R(,F(*Location,R ,F ) vided by any rural lands as Zr and Pr, and refer to
i P i 0 0
these as “generic” rural amenities (i.e., having the
and land in an agricultural use is not a necessary condi-
tion for providing these amenities). Thus, farmlands
(1d) Z1'GZ(R(,F(*R0,F0), provide Zf, Pf, Zr, and Pr, while other rural lands
(e.g., forests) provide only Zr and Pr.
subject to
Second, when V is not additively separable, the
marginal utility of rural amenity j (dV/dZ, or
(1e) M'∆F(C . j
F dV/dP) will depend on the levels of Z and P. In
j
other words, the marginal values of different rural
G is a vector of functions which determine the
P amenities are not static; they will depend on what
values of the sequence of P (where P is a vector of
i i amenities are currently available. For example,
prices); it depends on the expected, intertemporal
when Zf and Zr are not close substitutes, the ratio
distribution of farmland and other rural land types.
of the marginal utility from increasing generic rural
This distribution is influenced by the current distri-
amenities (dWTP/dZr) to the marginal utility of
bution of farmland (F) and other rural land types
0 increasing uniquely agricultural rural amenities
(R), and the farmland (F*) and other rural land (R*)
0 (dWTP/dZf) will diminish as R* increases. This has
currently enrolled in conservation and preservation
implications for the focus of farmland preservation
programs.6 It also depends on the distribution of
agencies. Specifically, if other protected rural lands
land uses relative to the individual’s residence
are providing an ample quantity of Zr, welfare-
(Location). Similarly, G is a vector of production
i Z maximizing farmland protection programs can be
functions for Z. It can also depend on R and F, but
expected to focus on protecting Zf. Conversely, if
does not depend on the location of an individual
R* is small, then welfare-maximizing farmland
residence. M is the agency’s budget, ∆F* denotes
protection programs can be expected to protect both
the farmlands protected in the current period, and
Zf and Zr.
C is the price per unit for these “protected” farms.
F Third, over the long term, the production of Zf
This model highlights the complexity of the
and Zr depends on the relative sizes of F* and R*.8
farmland preservation agency’s decision, involving
Consider Zr1, a generic open space amenity. Assume
expectations of utility, with the utility dependent on
farmland and other rural lands provide this open
space amenity equally well, and assume the produc-
tion of this open space amenity shows decreasing
5 This optimization problem is a simplification of an underlying maxi-
mization of a social welfare function, a function which may be nonlinear
in the contributions of individual WTP.
6 This formulation assumes the expected intertemporal distribution of 7 Note, this bifurcation is a simplification of a continuum, with differ-
rural land uses is conditioned on current land uses (R and F) and current ent rural amenities being more or less dependent on the presence of active
land protection (R* and F*). A more complete model 0could i0nclude infor- agriculture.
mation on expectations of future land use (such as predicted rates of pop- 8 This point can also apply to R and F, the current level of rural lands
0 0
ulation growth). and farmlands.
134 April 2003 Agricultural and Resource Economics Review
returns to scale (d2Zr1/d2R*< 0, and d2Zr1/d2F*< 0). the rights. The landowner retains all other rights to
Then, the long-run change in this open space the land. The government achieves the objective of
amenity due to an increase in protected farmland maintaining a rural land use, but does not obtain the
(dZr1/dF*) will be greater when the level of pro- capability to finely manage the land.
tected nonagricultural rural lands (R*) is smaller Increasingly, farmland protection agencies are
(d2Zr1/dF*dR* < 0). In contrast, consider Zf1, an ag- adopting PDR programs to permanently preserve
ricultural agrarian cultural heritage amenity. Since farmland. PDR program managers must also make
this is not supplied by R*, the long-run change in choices about which parcels of farmland best
this agrarian cultural heritage amenity due to an accomplish program goals. While PDR programs
increase in protected farmland (dZf1/dF*) will be are voluntary and participation will depend in part
independent of the current quantity of R*(d2Zf1/ on idiosyncratic differences among landowners,
dF*dR* = 0). the programs are typically oversubscribed, thereby
Fourth, the relative values of many types of rural allowing the program manager to choose which
amenities are a function of land use patterns and the parcels to preserve from the pool of applicants.
geographic distribution of the population. This is The conceptual model described by equations
captured in equations (1c) and (1d), where land use, (1a)S(1e) is complex and contains information
and the location of individuals relative to the amen- requirements that are not inconsequential for any
ities produced by different land uses, will influence farmland protection program. In practice, voluntary
the current prices and quantities of different rural programs such as PDRs are implemented using a
amenities. set of ranking criteria to choose among land parcels
with varying characteristics. These ranking criteria
reflect the agency’s best efforts to achieve program
Farmland Preservation Programs in Practice
goals, including the protection of rural amenities.
Government actions directed at preserving farm- The program manager’s problem is one of
lands take the form of both regulatory and volun- choosing parcels for preservation from a set of J
tary approaches. Many jurisdictions utilize zoning applicants (j = 1, ..., J), based on the highest ranked
regulations, both voluntary (e.g., agricultural parcels according to a given ranking scheme. Adapt-
districts) and involuntary (e.g., agricultural zoning). ing a simple, linear ranking mechanism used to
Zoning is often associated with differential assess- model other conservation programs (Cattaneo et al.,
ment, which sets property taxes based on current 2002), the decision can be characterized as:
land uses, rather than “best and highest use” values.
K
History has shown that zoning and differential
assessment rules have been modified when devel- (2a) MaxjBj jαkxij ,
Bj j0J k'1
opment pressures become sufficiently strong. Thus,
subject to
they have not proven to be a dependable means of
maintaining rural land uses over the long run. J
Outright government ownership is also a means of (2b) jBj(Ej#M,
j'1
preserving land uses, and is a common strategy for
providing amenities for which public access is where B = 1 if a parcel is chosen for preserva-
j
essential—such as the establishment of parks as a tion, α is the weight assigned to the kth objective
i
means of providing outdoor recreational experi- (k = 1, ..., K), and x is the jth parcel’s (or land-
ij
ences. However, government purchase of farmland owner’s) characteristic being weighted. The pro-
when it is maintained in that use is rare. gram manager chooses parcels subject to a budget
An approach in which the land is left in private constraint, where E is the cost of purchasing the
j
ownership and generally does not allow public easement (development rights) on the jth parcel,
access includes government programs that purchase and M is the amount of government funding avail-
partial interests in the land. Among these are the able for easement purchases.
Purchase of Development Rights (PDR) programs, Essentially, the weights (α) capture the contri-
k
which involve a voluntary agreement by the land- butions to social welfare from the various land
owner to forego development of the land. In PDR characteristics. They are expressions of the relative
programs, a government agency typically purchases preferences for different characteristics and amen-
all of the development rights, the rights are retired, ities associated with agricultural land; they may
and the landowner is paid a lump sum amount for also capture the relative scarcity of particular farm-
Nickerson and Hellerstein Protecting Rural Amenities Through Farmland Preservation Programs 135
land attributes. To the extent equations (2a) and mation on the types of objectives and amenities
(2b) are an accurate “reduced form” for equations sought for preservation through these programs, but
(1), optimizing these equations (given a set of did not reveal the relative importance of these
offered parcels) will optimize social welfare. goals. Second, a comparative analysis was con-
Just as farmland protection program objectives ducted of ranking criteria used in several state and
are likely to vary spatially across the country, we local PDR programs.
can expect the α’s to vary across preservation pro- Our goal in this part of the analysis was to
k
grams. In the next section, we investigate the factors discern how the priorities for preserving parcel
contributing to variation in these weights, and the characteristics, and weights assigned to the
implications for the protection of amenities and the priorities, vary across programs. The weights can
design of farmland preservation programs. be interpreted as indirect measures of the relative
importance of protecting the targeted parcel char-
acteristics and the underlying rural amenities
Farmland Protection Legislation and
associated with them. In the remainder of this
PDR Ranking Criteria as Data
section, the data and approaches used in the
Creating farmland preservation programs involves analyses are described.
familiar processes of government: passing enabling
legislation, securing funding from general revenues Enabling Legislation: Data and
or from a dedicated stream, and allocating funding Research Approach
through a bureaucracy. When these processes are
motivated by consideration of an underlying model The enabling legislation of farmland preservation
such as in equations (1), they can offer evidence as programs often contains statements relating to
to what objectives and amenities produced by farm- purpose or mission. We analyzed these mission
land are considered the most important to preserve. statements for a broad set of programs related to
Examination of PDR ranking mechanisms can agricultural land preservation. These programs
reveal the relative importance of protecting differ- include laws that establish agricultural districts,
ent parcel characteristics and their associated rural agricultural protection zoning, comprehensive
amenities. Although not motivated exclusively by growth management, conservation easements (such
public preferences for amenities, each of the steps as PDRs), differential assessment, and right to farm.
that occur in the creation and implementation of Using the American Farmland Trust’s (AFT’s)
farmland protection programs reveals something online links (AFT, 2000), or State sources for
about which amenities are considered to be the most codes, the purpose and findings clauses embedded
important to protect. Hence, we examine these pro- in these sets of State Codes were analyzed, and the
grams for evidence as to what rural amenities are occurrences of key phrases referring to specific
most likely to be protected, and how these vary as rural amenities and objectives were identified. The
socioeconomic and geophysical factors vary.9,10 initial step was to review the state-level laws
We analyzed two strands of evidence provided pertaining to farmland preservation in the 48 con-
by government programs designed to protect farm- tiguous states. This process yielded a large number
lands. First a content analysis was conducted of the of “catch phrases” relating to an underlying core set
enabling legislation of state-level agricultural land of objectives.
protection programs. This approach yielded infor- To synthesize this information, each phrase was
categorized on the basis of an identified objective.
Based on a literature review and on our reading of
9 A limitation of using legislation as evidence of importance of amen- the enabling legislation, we developed a list of five
ities is that enactment of legislation is sensitive to both the demand for
broad categories of objectives: (a) orderly devel-
rural amenities and to the supply of rural lands—i.e., in regions where
farmland is abundant, there is less need for farmland protection legis- opment, (b) food security, (c) local economy, (d)
lation even if the population of these regions has a high demand for
environmental services, and (e) protection of rural
farming-related amenities.
10 We note this analysis is akin to imputing shadow values by examin- amenities. These objectives are described in table 1,
ing the requirements of regulations, such as using shadow values from panel A.
cost functions due to pollution regulation. As with many such exercises,
Because we are specifically interested in infor-
there is always the issue of putting the cart before the horse—the analyst,
wishing to provide information for policy makers, merely echoes back the mation about which amenities are sought for pro-
current guesses of these self-same policy makers. However, to the extent tection, the four underlying components comprising
that objectification of observed practice can illuminate, this sort of ex
post analysis can be helpful. the “amenities” category are likewise considered,
136 April 2003 Agricultural and Resource Economics Review
Table 1. Legislative Language of Farmland Preservation Programs: Description of Objectives and
Description of Amenities Sought for Protection
A. OBJECTIVES OF FARMLAND PRESERVATION PROGRAMS
No. of States
Objective Description Citing Category
Orderly Development Orderly development of rural land, low density, physical space,
lower public utility provision costs, prevention of sprawl 18
Food Security Local and/or national food security: quantity or quality 30
Local Economy Preserve local agricultural/timber economy, agricultural or
timber jobs, other natural resource economies and jobs 24
Environmental Services Pollution reduction, groundwater recharge, flood control, water
quality/quantity, air quality 29
Protection of Rural Amenities Protection of open space, rural/agrarian character, wildlife
habitat, scenic beauty (see full description, panel B below) 36
B. AMENITIES SOUGHT FOR PROTECTION BY FARMLAND PRESERVATION PROGRAMS
No. of States
Amenity Description Citing Category
Open Space Usually visual, including prevention of or slowing of
development 31
Rural/Agrarian Character Agrarian cultural heritage, historic/nostalgic/unique terrain,
preserve farming/active agriculture/agricultural viability, live
rural way of life/sense of community/sense of place 31
Wildlife Habitat Includes wildlife habitat and natural areas 24
Scenic Beauty Aesthetics: usually visual, including scenic beauty, viewing
wildlife or farm activities 30
Note: Number of States’ legislation reviewed = 48.
and are detailed in panel B of table 1: (a) open parcels for preservation on the basis of least cost.
space, (b) rural/agrarian character, (c) wildlife habi- Other programs, including the Massachusetts State
tat, and (d) scenic beauty.11 program, Vermont State program, and the Sonoma
County, California, county program identify
preservation priorities but are excluded from the
Ranking Criteria: Description of Data
analysis because the criteria for prioritizing
In the second stage of the analysis, we analyzed the purchases is not clearly designated with a point
ranking criteria adopted by 27 separate state and system.
county PDR program agencies. PDR programs were Because state and local governments in the North-
included which (a) currently use point systems to east have been the first to adopt PDR programs and
objectively rank parcels; (b) were established tend to have the most active programs, observations
programs with a significant history of easement from programs in this region represent about 88%
purchases; and (c) were oversubscribed, so that the of our sample. Consequently, this analysis will be
rankings were actually used to select or reject par- most representative of the Northeast. We recognize
cels for preservation (AFT, 2002a,b). Several PDR the Northeast states may have priorities for rural
programs, such as the Delaware State program, are land protection which are systematically different
not included because they currently prioritize from the rest of the nation because they are the
most developed, may have different settlement
patterns and population demographics, and because
11 We evaluated the legislation by using a “yes/no” indicator if at least
they have the broadest and oldest set of farmland
one of a State’s farmland protection laws mentioned that particular out-
put. We also evaluated the legislation using a weighted classification, preservation programs. However, in many ways,
which assigned higher scores when the legislation contained more lan-
the Northeast may be a bellwether for other rapidly
guage about a given output. The conclusions drawn were not significantly
different from those reported here. growing regions.
Nickerson and Hellerstein Protecting Rural Amenities Through Farmland Preservation Programs 137
Table 2. Factors Considered in PDR Program Ranking Criteria and Descriptive Statistics
Factor for Which Mean
Points Are Awarded Description (Std. Dev.) Minimum Maximum
Soil Productivity Soil productivity or capability, percent tillable, use of 36.296 0 75.000
land for crops and pasture (19.207)
Farm Importance Owner operated, history of family farming, important 9.519 0 50.000
to agricultural community, farm capital improvements, (11.369)
specialty farm/unique production
Development Pressure Significant nonagricultural use nearby, near/in water 12.444 0 40.000
and sewer service areas, minimal septic limitations (9.394)
Road Access/Frontage Parcel has public road frontage 3.611 0 19.000
(5.565)
Environmental Significance Environmental, historic, or scenic importance; water 5.685 0 45.000
and soil conservation plans (9.225)
Parcel Size/Contiguity Parcel size, adjacent or near easements/districts/ 25.630 0 60.000
agricultural security areas/permanent open space, in (15.719)
agriculturally zoned area
Other Local government support, relative best buy, special 6.815 0 35.000
conditions (9.684)
Notes: N = 27 PDR programs. Descriptive statistics are based on the percentage of total points allocated to the identified factor.
The ranking criteria for all the programs taken the percentage change in the levels of PopDen,
together considered 36 different characteristics, %Farm, and %Prime, respectively, between 1987
which were related to several broad factors: soil pro- and 1997, using Census of Population and NRI data
ductivity, farm importance, development pressure, for those years. A lack of data precludes consider-
road access/frontage, environmental significance, ation of the effect on the weights in farmland pro-
parcel size and contiguity, and other considerations. tection program ranking criteria by acres preserved,
Table 2 summarizes these data.12 To facilitate com- or by types of land preserved, in other rural land
parison of the ranking systems, the descriptive protection programs.
statistics are reported in terms of percentage of total
points allocated to the various priorities.
Results and Discussion
We investigated whether underlying factors,
related to the demand and supply for farmland Enabling Legislation Analysis
characteristics, help explain the observed weights
given to the different priorities. Population density Map 1 graphically displays the states that identified
(PopDen) captures “demand” effects, while the each of the objectives in their farmland preservation
percentage of land that is farmed (%Farm) and enabling legislation. This map overlays the objec-
percentage of land that is prime land (%Prime) tives on two base layers. The first base layer (solid
in the jurisdiction capture “supply” effects. The shading) is the location of cropland and grassland
PopDen variable is derived from the Census of pasture and range, from the 1990 National Land
Population intercensal estimates for 1997 (U.S. Cover Database. The second base layer (hatched
Department of Commerce, Bureau of the Census, areas) identifies areas of the country subject to urban
2000), and the %Farm and %Prime variables are influence. The extent of urban influence was
obtained from the Natural Resource Conservation determined using a gravity model and 1990 Census
Services’s 1997 National Resource Inventory (NRI) of Population data by block group.13 Although land
(U.S. Department of Agriculture). We also looked
at trends in these three variables. The variables
PopDenChg, %FarmChg, and %PrimeChg capture 13 An urban influence index, essentially a measure of urban proximity,
was derived for the entire United States using Census block population
data and GIS-based statistical smoothing techniques. This measure is
derived from a “gravity” model of urban development, and increases as
12 Points assigned to use of land for specialty crops could arguably be nearby population increases and/or as distance from the parcel to popula-
allocated to the “soil productivity” category, rather than the “farm impor- tion decreases. The index used population within a 50-mile radius of each
tance” category as we have done. Either way these points are allocated, parcel, and a linear (rather than a squared) inverse distance weighting (see
the general conclusions discussed in this paper remain the same. Barnard, 2000, for further details.)