Flood hazard information disclosure by realtors

[From the U.S. Government Printing Office, www.gpo.gov]

Natural

Hazard

Research

FLOOD HAZARD INFORMATION
DISCLOSURE BY REALTORS

John A Cross
Department of Geography and Urban Studies
University of Wisconsin-Oshkosh

May, 1985

Working Paper #52
HC
79
.D45
C76
1985

This material is based upon work supported by the National Science
Foundation under Grant No. CEE-8211441.
Any opinions, findings, conclusions, or recommendations expressed in
this paper are those of the author and do not necessarily reflect the

views of the National Science Foundation.

SUMMARY i

Research in California has shown that mandatory disclosure of earth-

quake hazard information has had very little impact upon home purchases

in danger areas. This study looks at a flood hazard area in the Lower

Florida Keys to see if there is a similar phenomenon. Questionnaires

were sent to realtors, residential home buyers, and mobile home buyers

within the study area (between Big Pine Key on the north and Saddlebunch

I ~~Keys on the south) between January of 1980 and August of 1983. The

response rate was 67% for realtors and residential real estate purchasers

and 64% for mobile home buyers. The results are very similar to those in

* ~~California: although the disclosures do have a positive influence upon

the acquisition of flood insurance, they have not diminished the growth

U ~~of population in flood hazard areas of the Florida Keys.

V~~~~~~~~~~~~~

PREFACE

This paper is one in a series on research in progress in the field of
human adjustments to natural hazards. It is intended that these papers be

used as working documents by those directly involved in hazard research,

and as information papers by the larger circle of interested persons. The

series was started with funds from the National Science Foundation to theI

University of Colorado and Clark University, but it is now on a self-

supporting basis. Authorship of the papers is not necessarily confined to

those working at these institutions.

Further information about the research program is available from the
following:I

William E. Riebsame
Institute of Behavioral Science #6
University of ColoradoI
Boulder, Colorado 80309

Robert W. KatesI
Graduate School of Geography
Clark University
Worcester, Massachusetts 01610I
Ian Burton
Institute for Environmental Studies
University of TorontoI
Toronto, Canada M5S 1A4

Requests for copies of these papers and correspondence relatingI

directly thereto should be addressed to Boulder. In order to defray

production costs, there is a charge of $3.00 per publication on a sub-
scription basis, or $4.50 per copy when ordered singly.

I~~~~~~~~~~~~~~~~~~~~~~~~~~~~

TABLE OF CONTENTS

List of Tables. ........................vii
I ~~~List of Figures. .......................Vii

Introduction .1..............
The Lower Florida Keys Study. ..................3

* ~~~Realtors as Disclosure Agents.................. 5

Disclosures and Home Buyer Characteristics .............13

I ~~~Effects of Hazard Disclosures. .................17

I ~~~~The Realtors' Perspective. .................17

The Home Buyers' Perspective. ...............19

* ~~~~Residents' Perceptions of Hurricane Threat. ........27

Importance of a Home's Storm Safety. ..............28

I ~~~Disclosures and the Purchase of Mobile Home Residences. .....29

* ~~~Conclusions. ..........................33
References .............................38

viI

LIST OF TABLES

Table

I Proportions of Home Buyers Receiving Hurricane Flood Zone
Information......................... 7

2 Hurricane Hazard Disclosure and Method of Home Acquisition .7

3 Methods.Realtors Claim to Utilize to Inform ProspectiveI
Customers that Homes are Located in Hurricane Zones..... 8

4 Methods Utilized to Provide Home Buyers with Hurricane
Flood Information, 1977-1983. ...............10

5 Typical Questions Asked About Hurricanes Reported by
Realtors. .........................11I

6 Post-1980 Hurricane Hazard Disclosure and the
Homes' Appraised Building Value (1983). ..........13I
7 Post-1980 Hurricane Hazard Disclosure and the
Acquisition of Homes On and Off Big Pine Key. .......14

8 Methods of Hurricane Hazard Disclosure Used with
Post-1980 Newcomer Buyers and Post-1980 Resident Buyers . . . 15

9 Association Between Hurricane Hazard Disclosure and the
Importance of "Safety of Home During a Hurricane" in the
Selection of the Home. ...................20I
10 Association Between Hurricane Hazard Disclosure and the
Importance of the "Vulnerability of Homesite to
Storm Flooding" in the Selection of the Home. .......20
11 Hurricane Flood Zone Disclosure Methods and Their Effects
Upon the Post-1980 Buyers' Home Selection Process. .....23
12 Association Between Hurricane Hazard Disclosure and the
Acquisition of Flood-and Windstorm Insurance. .......25
13 Association Between Hurricane Hazard Information Disclosure
and the Perceptions of Post-1980 Buyers of the Vulnerabili-
ty of the Lower Florida Keys to Hurricane Damage. .....27

LIST OF FIGURESI
Figure

I Map of Florida Keys Showing Study Area........... 4

INTRODUCTION

Realtors and mortgage bankers are increasingly being given the task

I ~~of providing prospective home buyers with information concerning the

risks involved in moving into various hazard zones. In California, state

law requires-that realtors inform home purchasers that homes are located

near fault lines. Dade County, Florida, requires buyers of lands in the

Everglades to sign written statements indicating they are aware they are

I ~~purchasing swampland. Within coastal flood zones or on river flood

plains, mortgage lenders and often realtors must inform customers that

homes are located within flood zones and that flood insurance may be

required. When interstate sales of lands covered by the Interstate Land

Sales Full Disclosure Act are made, developers must file reports indi-

cating whether lands are subject to floods, hurricanes, or other hazards.

Within Flo~rida, as indicated by Baker and McPhee (1975, p. 85), it is

U ~~"mandatory for hazardous conditions (flood hazards) extant on the land to

be specified in the [subdivision] reports."

Even where laws do not explicitly require property sellers or real

estate agents and brokers to disclose hazards which may affect a proper-

ty, the courts may hold them liable for failure to do so. For example,

I ~~Platt (1983, p. 302) cites a court case which found that where "sellers

* ~~'held themselves out as highly skilled and competent land and resort

developers,' they were subject to a duty to investigate and disclose

potential erosion hazards." Slosson and Havens (1984), referring to

several California court decisions, indicate that the liability of real

I ~~estate salespersons has been extended even further, and that brokers not

only have the duty of informing potential buyers about any geologic

hazards, but also the responsibility of inspecting the property andI

seeking public data concerning any hazards which may be present.

Disclosure of such hazard information not only has the purpose of

protecting the property buyer by its "truth-in-selling'" character, but as

Simpson and Riehl (1981, P. 286) comment, it may "impose upon the de-

veloper and user of coastal property the reality of natural hazards." In

a report prepared for the Florida Bureau of Disaster Preparedness, Wilson

and colleagues (1980, p. 141) describe how subdivision regulations "can
assure... .that flood hazards are disclosed before any sale of land is made

to prevent victimization and fraud." They also discuss the benefits of

requiring the seller to inform the purchaser that property is in a flood

plain, and to disclose "the types and amounts of damages incurred through

the years from floods or other natural phenomena" (Wilson et al., 1980,

p. 180). Likewise, a recent Federal Emergency Management Agency publica-3

tion states, "Real estate agents could play an important role in inform-

ing prospective buyers about the vulnerability of coastal properties,

though they are unlikely to do so unless required by local ordinance"

(Field Associates, 1983, p. 26).

However, many questions remain concerning the effects of providing

home buyers with hazards information. While the effects of realtors upon

the home selection process in general have been reviewed in considerable

detail (see Palm, 1976 and 1981b for an excellent summary), their impacts

as purveyors of hazards information is less clear. Indeed, Baker andI

McPhee (1975, p. 106) comment, "It is known that real estate developers

often attempt to influence hazard zone land use decisions, but the extent

to which their influence is felt is not known." Research conducted in

1 ~~~~~~~~~~~~~~~~~~~~~~~~~~~3
California (Palmn, 1981a, 1981b) concerning the earthquake hazard indi-

I ~~cated that even man~datory disclosure of earthquake information by real-

tors to home buyers within "special studies zones" had little impact upon

home purchasing.

* ~~~~The effect of similar disclosures within flood hazard zones upon

home selection has not been well-studied. The flood hazard particularly

warrants study since property is often deliberately purchased within

coastal hazard zones to obtain the amenities of shoreline living, while

I ~~fault zone properties generally lack such special environmental attrac-

tions. Indeed, residents of coastal flood zones have been found to be

among the most educated concerning the presence of environmental hazards,

* ~~showing greater awareness of hazards than residents of river flood plains

(Burton and Kates, 1964). Further, since shorefront homes frequently are

I ~~second or vacation homes representing surplus capital (Mitchell, 1974, p.

20), it is possible that buyers' evaluations of the hazard might differ

from those which would be expressed if a primary residence were being

* ~~~purchased.

Thus, while the impact of earthquake information upon home sales has

I ~~been well-studied in California, the applicability of those findings to

the provision of other hazards information is questionable. This paper

explores the effects of flood zone disclosures by realtors within a

I -~~hurricane flood zone.

I ~~~~~~~~THE LOWER FLORIDA KEYS STUDY

* ~~~~The effects of hurricane hazard disclosures upon the home selection

process were examined from the perspectives of both realtors and home

buyers within the Lower Florida Keys. Data were collected concerning

whether realtors provided their customers with hurricane information and

the methods utilized to convey flood hazard data. During the 1983 hurri-

cane season questionnaires were sent to members of the local boards of

realtors, covering the area from Key West to Marathon, the largest

communities on either side of the study area. A smaller sample of real-

tors (N = 53), shown by the questionnaires to be actively selling resi-

dential real estate, were questioned in depth using an interview survey

form. All persons who purchased houses within the study area (from Big

Pine Key to Saddlebunch Keys) (see Figure 1) between January 1980 and

August 1983, according to county property ownership records, were mailed

an eight-page questionnaire.

KC7.

THE.

FLORIDA KEYS

(~ Ci~ li WlnI Ic Kel

Key
LDo* Key
STUDY AREA . , ' , Key

~~~~~i Plate K ey
$edd11 Beech 'r~eys '~/$.egerloel Key ':'h
K� .' '.W
M~~rqu~~odde DM)
Keys ,~~~~~~~~~~~~~":"-' f. .. 0

FIGURE I

MAP OF FLORIDA KEYS SHOWING STUDY AREA
~,x~ ""*~~~~~~~
~' - ~:'~~~~~

U ~~~~Three hundred and fifty-seven individuals returned questionnaires, a

response rate of 67%. Additionally, a small proportion of 1977', 1978,

and 1979 house purchasers were also sent questionnaires (145 were

* ~~~returned) so that any changes in the use of and effects of the hazard

disclosures could be discerned. Responses to a similar questionnaire

I ~~designed for~mobile home residents were received from 120 persons (67%

response rate) who purchased mobile homesites after January 1980.

Respondents (about 2%) who indicated they acquired their homes through

inheritance or gifts were excluded from statistical analysis. Although

data were collected from both buyers of houses and mobile homes, the

I ~~~primary focus of the analysis within this paper is upon the purcha sers of

* ~~~houses.

The Lower Florida Keys study area at present has over 5000 dwelling

units along with nearly 1000 mobile home or recreation vehicle spaces

which are seasonally occupied. Of the dwelling units, approximately 60%

I ~~are houses, with most of the remainder being mobile homes. The highest

* ~~local elevation--eight feet--is found in a small area of Big Pine Key;

approximately 96% of the study area is estimated by the Florida Coastal

Coordinating Council (1974, p. 1) to be situated below the five-foot

contour. Because the minimum base flood level of the Federal Insurance

I ~~Rate Maps for the study area is nine feet, the entire area could be

inundated during a 100-year storm. Within the velocity zones nearer the

shore, flooding of 13 feet is possible.

REALTORS AS DISCLOSURE AGENTS

I ~~~~Prospective home owners are not routinely advised about the threat

of hurricane flooding by many realtors, even though virtually the entire

study area lies within designated flood zones. (Again, only 4% of theI

land in the Lower Florida Keys is more than five feet above sea level.)

Only 52% of the new home owners (those who obtained houses since January,

1980) responded affirmatively to the question, "When you purchased your

home or homesite, did your realtor (or previous owner) inform you that

your home was in a hurricane flood zone or that you might need flood

insurance?" On the other hand, two-thirds of the area realtors claim

they told prospective home buyers that the property was located within a
hurricane flood zone before a deal was closed, while another 19% indica-

ted that they advised that flood insurance might be required, although

they did not specifically discuss hurricanes.

The proportion of home purchasers reporting that their realtor or

the previous home owner informed them that their home or homesite was

either in a flood zone or needed flood insurance has nevertheless climbed

steadily over the past few years. Thirty-five percent of home buyers

were so cautioned in 1978 (Table 1); this had risen to 66% by 1982,

although it dropped to 55% in 1983. One quarter of the new owners built

their own homes upon the homesites which they purchased. Of those whoI

purchased already constructed homes, approximately three-quarters bought

their property through a realtor. Of those individuals who dealt direct-

ly with the previous owner since January of 1980, 52.6% indicated that

the seller informed-them about the property's coastal flood vulnerability

compared with 58.5% of those buyers who dealt with a realtor (Table 2)--I

*hardly a significant difference.

Undoubtedly, many persons who purchased their homes directly from

the previous owner received hurricane information from realtors contacted

while considering other properties. Therefore, it is possible that these

TABLE 1

PROPORTIONS OF HOME BUYERS
RECEIVING HURRICANE FLOOD ZONE INFORMATION

Information No
Year Home Acquired: Received Information

Before 1970 20.4 % 79.6 %

1970 through 1977 32.0 % 68.0 %

1978 34.9 % 65.1 %

1979 39.2 % 60.8 %

1980 45.7 '. 54.3 %

1981 48.5 7% 51.5 %

1982 65.5 % 34.5 %

1983 55.1 % 44.9 %'

Chi-square=37.27, significance=.0000

TABLE 2

HURRICANE HAZARD DISCLOSURE AND METHOD OF HOME ACQUISITION

Hazard Hazard
How did you obtain your Disclosure Disclosure
Lower Florida Keys home? Given Not Given

Bought it through a realtor 96 68
(58.5%) (41.5%)

Bought it directly from
previous owner 40 36
(52.6%.) (47.4%)

Built the house with own labor 7 25
(21.9%) (78.1%)

Had home built by contractor 16 34
(32.0%) (68.0'.)

Chi-Square = 21.550, 3 degrees of freedom, significance = 0.0001.

figures do not realistically reflect differences which might exist, since

newly arrived home seekers were more likely to deal through a realtor

than the already-resident buyer. In contrast, only 28.1% of the resi-

dents who purchased a homesite since 1980 and subsequently constructed a

house reported receiving information at the time they acquired their

homesite.

The method of hazard disclosure as reported by the realtors was

highly variable (Table 3). Only 27% of the realtors interviewed indica-

ted that they or their offices had any set policy concerning how the

disclosures were made. Typically, the realtor stated that his or her

customers were informed about the coastal flood hazard while discussing

TABLE 3

METHODS REALTORS CLAIM TO UTILIZE TO INFORM PROSPECTIVE
CUSTOMERS THAT HOMES ARE LOCATED IN HURRICANE FLOOD ZONES

"How do you usually disclose Percent of Realtors
this information?" Making Each Response

Conversation about insurance in general . . . .32

Written disclosures at or before closing . . .. . . . .

Discussion of insurance with financing . ..... . 26 -

Comes up in general conversation . . . . 26 %

Conversation while showing homes and area .. . .. 12 %

Specific mention of flood levels and hurricanes . 10 %

Contrast of local problems with other areas . . 4 %

Note: Responses do not add to 100 percent since some realtors
gave several responses to the open-ended question.

general home owners' insurance needs (32%) or while discussing insurance

requirements for financing (26%). Less frequently, the realtor pointed

out insurance needs while showing homes (12%), or by specifically dis-

cussing flood levels or information from "the National Hurricane Center"

(19%). Likewise, only 10% of the realtors claimed to have made use of

flood zone maps in their discussions with customers. Many realtors

stated that they themselves had never seen the maps. Most of these dis-

closures were made verbally; only 30% of the realtors who inform their

customers stated that they made written disclosures at or before closing.

It is interesting to note that although the current Marathon and

Lower Keys Board of Realtors' disclosure form explicitly mentions flood

insurance, many realtors are using older versions which only disclose the

cost of "hazard insurance." Nevertheless, only 10% of the realtors

interviewed indicated they felt that any "requirements that realtors

should inform prospective buyers that property is located in a flood

zone" were unfavorable. On the other hand, only 32% of the realtors

expressed favorable attitudes towards the accuracy of the Federal

Insurance Administration's flood hazard maps.

One-third of this decade's home buyers who reported receiving hurri-

cane hazard information indicated that it was provided during discussions

about hurricanes when homes were being shown. Another 14% receiving

disclosures indicated that their realtor "talked about hurricanes during

[their] first office visit." However, as can be seen in Table 4, this

early disclosure was more common during the 1970s. The remaining in-

formed home buyers were usually told about hurricane problems just before

closing with a written disclosure form, or at the time financing was

arranged.

TABLE 4

METHODS UTILIZED TO PROVIDE HOME BUYERS
WITH HURRICANE FLOOD INFORMATION 1977--193

"How did your realtor (or previous Year Home Purchased
owner) inform you that your home was Average Average
in a hurricane flood zone or that 1977-1979 1980 1981 1982 1983 1980-1983
you might need flood insurance?" * *

Talked about hurricanes during
first office visit . . . . . . .... 24 ' 17 '. 13 17 % 8 % 14 %7

Talked about hurricanes when
homes were shown. . . . . . . ... 37 '. 43 32 % 26b % 38 % 34 %

Talked about hurricane problems
just before closing. . . . . . .. 10 '. 1 % 13 14 % 17 % 14 '.

Included information on written
disclosure form .......... 1 % 7 % 13 .% 17 % 13 I, 12 %'

Flood insurance requirements mentioned
when financing was discussed** ... 17 % 10 "I 13% 9 % 4 %7 9 '

Miscellaneous other methods**
2 %. 12 % 17 % 17 % 21 % 16 %

* Averages are weighted by the number of real estate sales in each year.
t* These responses were listed by home buyers who specified an
"Other" response.

The realtors reported that relatively few customers queried them

concerning hurricanes. Indeed, when the realtors were asked, "What

proportion of your clients have questioned you about the hurricane safety

of various houses or homesites?," 38% stated that less than a fifth of

the new home purchasers asked such questions, while another 34% of the

realtors reported that between one-fifth and three-fifths of their

customers had questions. When asked to typify the questions which they

were asked, the realtors' responses (Table 5) indicated considerable

TABLE 5

TYPICAL QUESTIONS ASKED ABOUT HURRICANES REPORTED BY REALTORS

Questions which realtors reported were Percent of Realtors
most frequently asked concerning Making Each Response
potential hurricane problems:

THE PHYSICAL HURRICANE HAZARD
What happens?, What are flood heights?,
How often are hurricanes? 51

PAST HURRICANE EXPERIENCES
What has happened in the past?, Have
there been any recent storms? 37 7.

RELATION BETWEEN BUYER AND HAZARD
How would the customer or his/her home
be affected? 45 %

THE REALTOR'S EXPERIENCES
What are the realtor's concerns and the
realtor's experiences with storms? 37 7.

MITIGATION BEHAVIOR
Does one evacuate? Should insurance be
purchased? 24 %

MISCELLANEOUS
Casual questions and jokes. 6 %

Note: Percentage figures exceed 100 percent because most realtors
gave more than one response.

interest, if questions were asked at all, in recent storm experiences andI

storm frequency. Since the area has gone unscathed since 1966, and

because nearly three-quarters of the real estate salespersons themselves

moved to the Keys since then, the accuracy of the realtors' answers is

highly questionable.

These findings suggest that either many new home owners already hadI

ideas about the potential for hurricane destruction within the area or

that they are poorly informed about the hurricane hazard. New home

owners appear generally cognizant of the risk, with nearly two-thirds

expecting a damaging hurricane within the next decade. Indeed, 77% of

the realtors indicated that over four-fifths of their customers wereI

aware that the area is subject to hurricanes. Nevertheless, there are

many prospective home buyers who wished their realtor had discussed

hurricane problems with them in more detail. Additional information was

desired by 32% of those customers not receiving hurricane information,

while only 14% receiving information wished for more details. Still, the

majority of both groups were satisfied with the information (or lack

the'reof) which they had received.

Although realtors indicated that the majority of prospective

customers realized that hurricanes had struck the area in the past and

may strike again, these customers had little knowledge about the exis-

tence of flood insurance and mortgage insurance requirements. For

example, 72% of the realtors reported that less than three-fifths of
their customers were aware of the existence of federal flood insurance

when they first contacted their realtor. In addition, 58% of the real-

tors interviewed indicated that less than one in three customers were

aware that flood insurance was required to obtain a home mortgage.

DISCLOSURES AND HOME BUYER CHARACTERISTICS

Disclosures were most likely to be given to, or at least remembered

by, purchasers of moderately priced housing (Table 6). Those buying

houses whose appraised building value (excluding homesite) according to

Monroe County property assessment records was below $20,000 were the

least likely to receive hurricane information from their realtor or

previous owner. Persons buying houses on Big Pine Key were significantly

less likely to receive hurricane flood information that purchasers on all

the other islands of the Lower Florida Keys (Table 7), probably reflect-

ing the fact that some homesites are over a half mile from the coast.

However, even those areas are in "A Zones," which would experience waters

several feet deep during a 100-year storm.

TABLE 6

POST-1980 HURRICANE HAZARD DISCLOSURE
AND THE HOMES' APPRAISED BUILDING VALUE (1983)

Hazard Hazard
Appraised Value of Disclosure Disclosure
Residence Building Given Not Given
(excluding homesite)

Less than $2,OC)0 12 24
(33.3%.) (66.7%.)

$20,001 through $40,000 54 58
(48.2%I (51.87.%)

$40,001 through $60,000 67 44
(60..4%) (39.6%.)

More than $60,000C 24 28
(46. 2.) (53.8/.)

Chi-Square = 9.188, 3 degrees of freedom, significance = 0.0269.

TABLE 7

POST-1980 HURRICANE HAZARD DISCLOSURE
AND THE ACQUISITION OF HOMES ON AND OFF BIG PINE KEY

Hazard Hazard
Home Located: Disclosure Disclosure
Given Not Given

Big Pine Key 48 73
(39.7%.) (60.37%)

Saddlebunch, Sugarloaf, 111 90
Cudjoe, Summerland, Ramrod, (55.7.2%) (44.8'%.)
Little Torch Keys

Chi-Square = 6.7C02 (after Yates Correction), I degree of freedom,
Significance = 0.0096

Note: The other keys were grouped so that cell sizes would be
suitable for statistical testing. All the other keys, however,
had higher rates of disclosure than Big Pine Key.

New home owners who were also new residents to the area were also

more likely to report disclosures. While 55% of the newcomers purchasing

houses or homesites upon which they built since 1980 were informed about

the hurricane flood hazard or the need for flood insurance, only 44% of

individuals who had lived in the Keys for at least a year before they

purchased their present home were so informed during the same time

period--a difference significant at the .10 level. Likewise, the manner

in which the realtor, or former home owner, provided the disclosure to

currently resident home seekers significantly differed (.05 level) from

that given the nonresident home buyer (Table 8). Newcomers who received

the disclosure were much more often informed while homes were actually

TABLE 8

METHODS OF HURRICANE HAZARD DISCLOSURE USED
WITH POST-1980 NEWCOMER BUYERS AND POST-1980 RESIDENT BUYERS

How did this person inform you? Newcomers Resident

Talked about hurricanes during 13 7
first office visit. (13.3.) (17.17.%)

Talked about hurricanes when 42 8
homes were shown. (42.9?.) (19.57.%)

Talked about hurricane problems 15 3
just before closing. (15.37.%) (7.3%.)

Included information on written 10 9
disclosure form. (10.2%.) (22.07.)

Other methods. 18 14
(18.47.) (34.17.)

Total 100C.07. 100. 0%

Chi-Square = 12.140, 4 degrees of freedom, significance = 0.0163.

being shown (42.9 %) than were other purchasers (19.5%).

It should be noted that a number of the already resident home buyers

volunteered on their questionnaires that their realtor did not have to

tell them about hurricanes because they were already aware. It is un-

clear whether the lower reported disclosure rate among resident home

buyers reflects diminished efforts at disclosure by realtors or home

sellers, or conversely, the reduced impact, and thus recall, of a dis-

closure by residents who are not hearing it for the first time. Even

though fewer resident home buyers reported the disclosures, only 19.8%

wished they had received more information concerning possible hurricane

problems, compared with 24.8% of the nonresident home buyers. Although

the differences in proportions are not statistically significant, they

suggest that pinpointing new residents for disclosures may occur.

The background of the newly arrived home seeker could have in-

fluenced whether or not the individual was informed about the potential

hurricane threat to the property being considered, but the data does not

strongly support this hypothesis. For example, of those post-1980 buy ersI

who were informed, 54.0% had previously lived in counties along the

Atlantic, Gulf, or Pacific Coasts, compared with 65.3% of those new home

buyers who failed to report such disclosures--a difference significant at

the .10 level. However, this difference can be attributed to differences

in disclosures noted between newly arrived home buyers and purchases byI

already resident individuals, since the differences within the two groups

on the basis of their former residence are not statistically significant.

Likewise, the proportion of residents receiving disclosures who came from

nonmetropolitan counties did not statistically differ from that reported

by those immigrating from urban areas.

Although the differences were again not statistically significant, a

greater proportion (47.3%) of the new home buyers not receiving hurricane

information had previously lived elsewhere in Florida before buying their

Florida Keys home, while 39.2% of those receiving disclosures were

Florida residents. These observations possibly hint at some biases whichI

the customers' former residence may have had upon their being given

information. On the other hand, there is no significant association

between whether or not a home purchaser received hurricane information

and his age, employment status, income, or education. Additionally, the

vulnerability of the purchased homesite to storm flooding was generally

not associated with whether or not the home buyer received hurricane

information from the realtor or previous owner at the time of the

purchase. For example, while 53.3% of the post-1980 purchasers of water-

front (shore or canal) homesites were given disclosures, compared with

41.8% of inland property purchasers, these differences were not signifi-

cant at the .10 level. Likewise, the proportion of home purchasers

receiving hurricane information was not statistically different between

buyers on the basis of the hurricane flood zone (velocity (V) versus

normal 100-year (A) flood zones), the base flood levels, or the elevation

(above or below five feet) of the homesites.

EFFECTS OF HAZARD DISCLOSURES

The Realtors' Perspective

The realtors interviewed were nearly evenly split in their opinions

about what effects their coastal hazard disclosures had upon their

customers' selection of house type (elevated on stilts versus ground

level). Indeed, 48% of the realtors interviewed reported that the dis-

closure aided in the selling of houses elevated on stilts, while 52%

disagreed. However, the flood information was viewed as having less

effect upon sales as a whole, or upon the geographic area of customer

interest. Only 18% of the realtors could recount a case where their

disclosure of flood hazard information had cost them any sales, and only

one in eight realtors could remember a case in which a customer had

looked elsewhere in the Keys following the realtors' hurricane dis-

closure. Several realtors mentioned that occasionally home seekers tried

to avoid certain islands such as Little Torch Key, preferring a larger

island such as Big Pine Key.

Still, house sales as a proportion of all residential sales wereI

statistically associated with whether realtors informed customers about

the flood hazard and whether they had a set policy about this disclosure.

For example, 62% of the realtors who had disclosure policies reported

that stilt houses accounted for over half of their residential sales,

while stilt houses provided over half the sales for only 29% of thoseI

realtors lacking disclosure policies. Likewise, 42% of the realtors who

claimed to routinely inform customers that homes are subject to hurricane

flooding indicated that stilt houses provided over half of their sales,

while only 29% of the realtors who did not inform customers reported as

high a proportion of stilt house sales.I

On the other hand, few of the customers' hurricane attitudes as

reported by the realtors could be related to the realtor disclosures or

disclosure policies. No statistically significant relationships exist

between realtors' informing customers and the attitudes, as perceived by

the realtors, which the home buyers had about stilt houses, minimum

elevation requirements, or flood insurance requirements. Likewise, those

realtors informing customers about the hurricane flood zones did not
report being asked about hurricanes by a larger proportion of their

customers than did realtors not routinely informing customers. However,

a weak statistical relationship does exist (at the .10 significance

level) between the existence of a set disclosure policy and the propor-

tion of customers who asked about hurricanes, with those realtors without

policies reporting that customers are more likely to ask questions.

The Home Buyers' Perspective

Only 12.4% of the new home owners informed about hurricanes indica-

ted that this disclosure influenced their selection of their homesite or

house type, although this proportion was 22.9% between 1977 and 1979.

I ~~Thus, it is not surprising that the individuals who received the hurri-

cane flood disclosures were statistically no more likely to purchase

stilt houses than were uninformed home buyers. Actually, 68.4% of post-

1980 home buyers who received hurricane disclosures purchased or built

stilt houses, compared with 72.4% of those that did not get the warning.

I ~~Likewise, a greater proportion of the home owners that got flood zone

disclosures purchased homesites along the shore or canals than did other

home buyers, although this difference was also not significant.
The home buyers receiving the hurricane disclosures did not statis-

tically differ from the other home purchasers with respect to the re-

I ~~ported importance of the "elevation of house above ground" in their home

selection. On the other hand, the importance of the "safety of home

during a hurricane" consideration was statistically related to whether or

* ~~not the home purchasers were informed about hurricane problems (Table 9).

Home buyers not receiving the flood hazard disclosure were nearly twice

* ~~as likely to have indicated that the hurricane safety was "not important

at all." However, home buyers who had received hurricane information

U ~~were slightly less likely to consider their home safety as *"very impor-

tant." This relationship was stronger among the newly arrived home

buyers than among the already resident home purchasers.

TABLE 9

ASSOCIATION BETWEEN HURRICANE HAZARD DISCLOSURE
AND THE IMPORTANCE OF "SAFETY OF HOME DURING A HURRICANE"
IN THE SELECTION OF THE HOME

Hazard Hazard
Importance of "Safety Disclosure Disclosure
of Home During A Given Not Given
Hurricane:"

Very Important 47 45
(51.1%) (48.9%)

Important 52 53
(49.5%) (50.5%.)

Slightly Important 36 17
(67.9%) (32.1%)

Not Important at All 15 24
(38.5%) (61.5%)

Chi-Square = 8.535, 3 degrees of freedom, significance = 0.0362.

TABLE 10

ASSOCIATION BETWEEN HURRICANE HAZARD DISCLOSURE
AND THE IMPORTANCE OF THE "VULNERABILITY OF HOMESITE TO STORM
FLOODING" IN THE SELECTION OF THE HOME

Hazard Hazard
Importance of "Vulnerability Disclosure Disclosure
of Homesite to Storm Flooding": Given Not Given

Very Important 38 49
(43.7%) (56.3%)

Important 54 54
(50.0%) (50.0%)

Slightly Important 39 21
(65.0%) (35.0%)

Not Important at All 18 21
(46.2%) (53.8%)

Chi-Square = 6.968, 3 degrees of freedom, significance = 0.0729.

I ~~~~Similarly, customers warned about hurricanes differed from those not

cautioned with respect to the "vulnerability of homesite to storm flood-

ing" in their home selection (Table 10). Although this relationship was

statistically weak (significance .08), purchasers not receiving hurricane

information were more likely to claim that their homesite vulnerability

I ~~was either "very important" or "not important at all." Of those

i ~~customers purchasing stilt houses rather than ground-level houses, those

who had received coastal flood hazard information from their realtors or

* ~~the former property owner were not significantly more likely to indicate

that flood protection was a very important factor in their choice. Thus,

I ~~~while several stati stically significant relationships were noted, for

* ~~most customers the hurricane flood hazard disclosure did not have a major

impact upon their home selection.

* ~~~~Whether or not home buyers were informed about the hurricane flood

hazard is only weakly associated (.10 significance level) with the resi-

I '~~dents' knowledge of their own home 's hazard vulnerability. Although

virtually all the residences within the Lower Florida Keys are in desig-

nated hurricane flood zones, home owners reporting that their realtors or

the home's seller discussed hurricanes were more aware of this fact. For

example, 62.6% of the informed home owners acknowledged their flood zone

I ~~~location, compared to 53.5 % of the others. (It is interesting to note

that this association is much stronger (.001 significance level) among

persons purchasing homes in 1977 through 1979).

* ~~~~Home owners' awareness that their residences were located in mapped

flood zones is generally unrelated to their overall perceptions of the

hurricane hazard. On the other hand, this knowledge about their flood

zone location is statistically related (at the 0.05 significance level)

to the residents' flood insurance coverage, their attitudes concerning

their ability to safely ride out a hurricane within their homes, and

their considerations of the vulnerability of their homesites to storm

flooding during the selection of their homes. Ironically, residents

aware that their homes were in a flood plain were less likely to claim

they could safely remain at home during a storm, yet they were less

likely to have considered the "vulnerability of their homesite to storm

flooding" as important when they selected their homes.

The methods utilized to inform prospective home buyers about the

hurricane threat had varying effects upon the home selection process

(Table 11). For example, discussions about hurricanes were twice as

likely to make a difference in a home buyer's choice of a house or home-

site (as claimed by the home purchaser) if they were made when homes were

shown (17.3%) compared with all other times (8.8%). Home purchasers who

reported that their realtors "talked about hurricanes during [their]

first office visit" were most likely to have considered as "very impor-

tant" in their selection of their homes the "elevation of house above

ground" (53%), the "safety of the home during a hurricane" (56%), and

the "vulnerability of homesite to storm flooding" (50%). These factors

were considered as "very important" to 29%, 26%, and 20%, respectively,

of the home buyers informed about the hurricane hazards at later times

during the home selection process.

The impact of informing customers about hurricanes early in the

house selection can also be shown by the finding that 55% of those

customers told about hurricanes just before closing or by a written

disclosure (usually at closing) bought stilt houses, while 72% of those

TABLE I1

HURRICANE FLOOD ZONE DISCLOSURE METHODS AND THEIR EFFECTS
UPON THE POST-1980 BUYER'S HOME SELECTION PROCESS

Proportion of Customers Considering Each Factor as
"Very Important" in their Home Selection
Disclosure Method
Elevation of House Safety of Home Vulnerability of
Realtor (or previous Above Ground During a Hurricane Homesite to
owner): Storm Flooding

Talked about hurricanes dur-
ing first office visit . . . 56 56. 50 '.

Talked about hurricanes when
homes were shown ....... .. 0 7 . 16 %

Talked about hurricane problems
just before closing . . . . . . . . 30 % 25%

Included information on written
disclosure form . . . . . . .. 7. .

Flood insurance requirements men-
tioned when financing was discussed -1 ' 3s %Z 15 7.

Miscellaneous other
. .~.~. ~. ~. . .. 22 % 5 %26 %

told about hurricanes during the first office visit or when property wasI

shown selected stilt houses. It should be mentioned, however, that

customers who received early disclosures, particularly during the first

office visit, were also significantly more likely (91.4%) to select

waterfront homesites than other buyers who received hurricane flood

information (78.9%).I

Flood and windstorm insurance was purchased significantly more often

by post-1980 home buyers who received disclosures concerning the hurri-

cane threat or the need for flood insurance from their realtor or the

previous home owner (Table 12). This statistical association exists for

both the newly arrived home seeker and the continuing resident who was

purchasing a new house. This relationship was expected, considering the

realtors' indications that they tended to point out hurricane problems

and the need for flood insurance when discussing financing and general

insurance needs. Indeed, the necessity of obtaining a mortgage gives

many home owners little choice in obtaining both windstorm and flood

insurance coverage. Nevertheless, of the post-1977 home buyers with

mortgages (as recorded in the Monroe County Clerk's Office) who were not

informed about hurricanes by their realtor or previous home seller, 5.6%

did not have flood insurance or were uncertain about having any coverage

(1.4%).

It is interesting to note that all of the above individuals lacking

flood insurance in violation of federal regulations have mortgages fromI

the same Monroe County financial institution. On the other hand, 100% of

those individuals with mortgages who were informed about the hurricane

threat had insurance. Even among those home owners lacking mortgages

(which would have the legal effect of requiring insurance), the provision

TABLE 12

ASSOCIATION BETWEEN HURRICANE HAZARD DISCLOSURE
AND THE ACQUISITION OF FLOOD AND WINDSTORM INSURANCE

Windstorm Insurance

Hazard Hazard
Di sc osure Di sc osure
Given Not Given

Windstorm Insurance Obtained 152 133
(97.4%) (84.2.%)

NO Windstorm Insurance Coverage 4 25
(2.6%) (15.8%)

Chi-Square = 14.918 (after Yates Correction), 1 degree of freedom,
Significance = 0.0001.

Flood Insurance

Flood Insurance Obtained 136 107
(86.1%) (67.7%)

NO Flood Insurance Coverage 22 51
(13.9%) (32.3%)

Chi-Square = 13.966 (after Yates Correction), 1 degree of freedom
Significance = 0.0002

of hurricane information at the time of the home sale has a significantI

association with insurance coverage. For example, 93.4% of home buyers

whose homes are mortgage free, but who were informed about the hurricane

hazard, have windstorm insurance; however, only 74.8% of those not in-

formed have such insurance, a difference significant at the 0.00005

level. Likewise, 77.7% of those home owners without mortgages who wereI

told about the hurricane threat while purchasing their homes have flood

insurance coverage, compared with 55.7% who were uninformed, another

highly significant relationship. Thus, even controlling for the insur-

ance requirements associated with home mortgages, the disclosure of

hurricane flood information by the realtor is strongly associated withI

the acquisition of flood insurance.

The post-1980 home buyers generally favor the eight-foot minimum

elevation requirements for new construction which were in effect when

their homes were purchased. Likewise, those favoring the use of higher

minimum elevations which include wave action (as shown on the FederalI

Insurance Rate Maps formally issued in December, 1983) outnumbered those

opposing their adoption by better than three to one. The association

between the home purchasers' favoring of the eight-foot requirements and

whether or not the nonresident home purchaser received a hurricane dis-

closure was not significant. However, significant differences (at theI

0.10 level) exist between those new resident home buyers who did and did

not receive hurricane information with respect to their attitudes towards

the adoption of the more stringent elevation levels and their attitudes

towards Monroe County's building code and its enforcement. For example,

68.8% of the home buyers who received hurricane information favored the

"use of these new [FIRM] maps to determine the minimum elevation levels

for new home construction," compared with 55.6% of the other home pur-

chasers. Conversely, home buyers who did not receive disclosures were

less than half as likely to feel that the present building code and its

enforcement was "too restrictive."

Residents' Perception of Hurricane Threat

The disclosures appear to have had little influence upon the home

owners' perceptions of the overall hurricane risk within the Keys.

Although informed customers are twice as likely to claim that the Lower

Florida Keys are more likely to experience hurricanes than other loca-

tions along the Gulf and Atlantic coasts of the United States, this

relationship is not statistically significant (Table 13). Indeed, the

majority of all recent home buyers feel that their vulnerability to

hurricanes is equal to that of other coastal locations. Similarly,

although slightly greater percentages of the survey respondents who were

TABLE 13

ASSOCIATION BETWEEN HURRICANE HAZARD INFORMATION DISCLOSURE
AND THE PERCEPTIONS OF POST-1980 BUYERS OF THE VULNERABILITY
OF THE LOWER FLORIDA KEYS TO HURRICANE DAMAGE

"Compared with other locations along the Gulf and Atlantic
Coasts of the U.S., do you believe a hurricane hitting the
Lower Florida Keys would be more, less, or equally likely
to cause property damage?"

MORE LESS EQUALLY

Hazard Disclosure 20 33 103
Given (12.8%) (21.2%) (66.0%)

Hazard Disclosure 10 32 104
NOT Given (6.87.) (21.9%) (71.2%)

Chi-Square 3. 026, 2 degrees of freedom, significance = 0.220.

informed about hurricanes by their realtor or previous owner feel that a

hurricane hitting the Lower Florida Keys would be more likely to cause

damage than at other coastal locations, these differences again lack

statistical significance. Likewise, in their evaluations of the hurri-

cane vulnerability of the Lower Florida Keys compared to Middle and Upper

Keys locations, of the anticipated damage to their homes if the center ofI

a hurricane were to pass over their area of the Florida Keys, and of the

changes in their hurricane awareness since they began living in the area,

no significant differences were discerned between those home owners who

received hurricane information from their realtor or home seller and

those who did not. Furthermore, the disclosures were not related toI

expected evacuation behavior in case of a hurricane warning. Thus, while

the provision of hurricane information to home purchasers may influence

certain aspects of their home acquisition process, it appears on the

whole to have been inconsequential in the development of their overall

perception of the hazard.

The methods utilized to disclose hurricane information appear more

closely related to differences in the new home owners' perception of the

area's hurricane risk. For example, 85% of those customers who were told

about hurricanes during their first office visit stated that a damaging

hurricane within the study area within the next decade was "likely" or

livery likely" compared with 57% of those receiving disclosures at other

times, and with 65% of those receiving no disclosures at all.I

IMPORTANCE OF A HOME'S STORM SAFETY

Realtors and new home owners disagree about the importance of a

dwelling's storm safety in the home selection process. For example, two-

thirds of the new home owners indicated that the home's storm safety was

"very important" or "somewhat important" in choosing a house. Converse-

ly, most realtors viewed this consideration as being of little impor-

tance--approximately equal to proximity of employment, a factor of

dubious importance to a population largely composed of retired persons or

vacationers. In fact, only 21.3% of house purchasers were concerned with

their location vis-a-vis employment.

New home owners are typically more concerned about the threat of

hurricane flooding than the area's realtors, just as they are more con-

cerned than most longtime residents. For example, hurricane flooding is

considered a "major problem" to 27% of the home owners and to 19% of the

realtors. Proportionately, twice as many realtors claim that hurricane

floods are "not a problem at all" as do new home owners.

DISCLOSURES AND THE PURCHASE OF MOBILE HOME RESIDENCES

Over one-third of the residences within the Lower Florida Keys are

mobile homes; realtors are less likely to be involved with mobile home

sales. Purchasers of mobile home residences were significantly more

likely (47.4% versus 31.0%) to have purchased the mobile homesite direct-

ly from the previous owner than purchasers of houses were likely to deal

directly with the former owner. The survey of realtors substantiates

these claims, with 55% claiming they sold no on-site mobile homes and an

additional 32% stating that such sales accounted for only 1%-5% of their

total residential sales. Such a low sales involvement was explained by

several realtors as due to most mobile home residences being sold among

residents or friends of residents living within the various mobile home

communities. Further, Florida's legal prohibition of realtors selling

30
mobile homes apart from their homesites was also frequently cited.I

Slightly less than half (46.5%) of the post-January 1980 mobile home3

buyers indicated that they obtained their mobile home and lot separately.

Typically, the purchaser of a mobile home homesite (with or without the

mobile home) spent less time searching than did the purchaser of a house

(64% of the former spending less than a month, compared with 51% ofI

house purchasers).

Slightly less than half of the mobile home purchasers reported that

their realtor, or the previous land owner, informed them that the proper-

ty was in a hurricane flood zone or that flood insurance might be needed,

a proportion not significantly differ ent from that reported by purchasersI

of houses. Likewise, the method of this disclosure was similar to that

received by home buyers. In addition, mobile home residents and house

residents appeared equally satisfied with the information they received.3

Although only a small proportion of either set of purchasers indicated

that the information made any difference in their choice of property,I

mobile home purchasers reported it only half as frequently.
Since April of 1977, a Monroe County ordinance (Ordinance 8-1977)

has required that variances be obtained from the Board of County Coin-

missioners for all new mobile homes placed at an elevation of less than

eight feet. The ordinance further stated:

If a mobile home is placed in an existing park or subdivision or
varianced lot, the Building Department shall not issue a BuildingI
Permit or Certificate of Occupancy unless and until the owner of the
mobile home or lessee, as the case may be, shows the Building
Official a recorded deed or unrecorded written lease wherein the
mobile home owner has the fact disclosed to him in said deed orI
lease that the mobile home is being located in a flood-prone area
and that an evacuation plan indicating vehicular access and an
escape route is filed with the Disaster Preparedness AuthoritiesI
(Monroe County 1977, pp. 4-5).

U ~~This ordinance, however, has been basically ignored. While mobile homes

have been added within the study area, no requests for mobile home vari-

ances had been filed with the Monroe County Building Department, which

had received only 16 such requests for the remainder of the county be-

tween January of 1980 and December of 1983. However, substantial changes

I ~~in personnel in the department were made in 1983, and the County Com-

mission enacted a new ordinance in November, 1983, placing stricter

elevation and variance requirements on mobile homes (Monroe County,

1983). Although not affecting real estate transactions during the time

frame of this study, it appears these regulations will have a greater

I ~~~impact in the future.

There is relatively little evidence that realtors have targeted any

particular group of mobile home seekers to receive the hurricane informa-

tion disclosures, except that a greater proportion of persons that ob-

tained their mobile home and homesite together reported receiving hurri-

I ~~~cane information (53.8%) than persons who purchased their lot separate

from their mobile home (38.0%). Yet, even this relationship was only

significant at the 0.134 level after Yates correction.

* ~~~~No significant differences existed between mobile home purchasers

receiving disclosures and those who did not with respect to their socio-

I ~~~economic characteristics (age, education, and income), or their previous

residence (census region, urban versus rural background, and coastal

versus inland location). It should be noted, however, that mobile home

buyers significantly differed from purchasers of houses on several of the-

variables. For example, only 19.5% of the mobile home purchasers were

I ~~~from the South Atlantic states, compared with 47.7% of the house buyers,

and they were less than half'as likely to have previously lived within a

coastal county. Likewise, as a whole,,mobile home buyers had less

income.

The hurricane flood disclosures had virtually no impact on the

mobile home purchasers' choice of a homesite--with only one of 54 post-

1980 buyers who received the hurricane information reporting that it

influenced his/her choice. Likewise, no statistical differences existed

between mobile home purchasers receiving and not receiving hurricane

information with respect to the importance in their home selection pro-

cess of the "elevation of the homesite," the "safety of home during a

hurricane," or the "vulnerability of homesite to storm flooding." How-

ever, it can be argued that because mobile home purchasers expect that

their homes would be destroyed by hurricane winds anyway, efforts to

obtain safer homesites would be futile. Indeed, 80.1% of the mobile home

purchasers expected that their homes would receive major damage or be

totally destroyed if the center of a hurricane were to pass over their

area of the Keys, with only 19.5% believing that flooding would be the

greatest cause of this damage. Thus, it appears that the perceived

futility of physically protecting mobile homes negates any impact which

the hurricane information might have on the choice of homesite.

The hurricane flood zone disclosures were much more strongly related

to decisions to purchase windstorm and flood insurance. Of those mobile

home purchasers told about the hazard or flood insurance, 90.4% carry

windstorm insurance, compared with 55.9% of the uninformed. Likewise,

82.7% of the buyers receiving hurricane disclosures purchased flood

insurance, compared with 44.1% of the other mobile home residents, a

difference significant at the 0.0001 level. Nevertheless, while nearly

two-thirds of all mobile home purchasers feel that the availability of

flood insurance will make it easier to sell their homes, no differences
existed between motile home buyers who did and did not receive the hurri-

cane information.

The provision of hurricane information had little impact upon the

mobile home buyers' perception of hurricanes, hurricane winds, or hurri-

cane waves and flooding as problems related to living in the Keys.

Likewise, no differences existed in the purchasers' perceptions of how

likely a damaging hurricane might be within the next decade. Mobile home

* ~~buyers with and without disclosures did not significantly differ in their

evaluations of the vulnerability of the Lower Florida Keys to hurricane

I ~~damage, as compared to other Gulf and Atlantic coast locations or the

Middle and Upper Florida Keys. However, purchasers receiving disclosures

were sign ificantly more likely to claim that the Keys were either more or

less likely to be struck by a hurricane (41.8%) than were other buyers,

80.3% of whom felt the Keys were equally vulnerable. Nevertheless, the

I ~~disclosure's impact upon perceptions appears minimal at best.

CONCLUSIONS

3 ~~~~While realtors may provide home buyers with coastal hazards informa-

tion, this information is not of great significance to most of their

I ~~customers when they select their homes. Indeed, only one-third of the

3 ~~home buyers who received disclosures considered realtors to be the best

source of information concerning minimum elevation requirements (100-year

flood levels), with county building and zoning officials (21%) and

insurance agents (14%) viewed by others as the best sources. Interest-

I ~~ingly, bankers were rarely (2.8%) viewed as a good information source,

even though it is their responsibility to assure that mortgaged homesI

carry flood insurance.

It is also possible that information about flood insurance may

reduce the overall concern of customers about property losses, since

buyers receiving disclosures more frequently located along the shore or

in ground-level houses. Many realtors admitted this w as the case. HalfI

of the realtors interviewed who sold houses within the Lower Florida Keys

affirmatively answered the question, "Have you found that the availabili-

ty of flood insurance makes your customers more likely to locate near the

water or on low-lying property?" Of course, it may also be argued that

because of these sites' greater vulnerability, realtors made greaterI

efforts to warn their purchasers.

Disclosure of hurricane flood zone information by realtors was

strongly associated with the home owners' acquisition of both windstorm

and flood insurance, even among residents lacking mortgages. Even though

the effects of the disclosures were not always apparent, it appears thatI

the disclosure of coastal flood hazard information by realtors could

indeed raise customer awareness and contribute to hazard avoidance be-

havior if the information is provided early in the home selection pro-

cess. However, the proportion of home buyers who received information

about hurricanes early in the home selection process appears to be drop-I

ping. Disclosures at the time of closing may fulfill legal obligations,

but are unlikely to alter a customer's selection. The disclosures given

at the time homes are selected have had no apparent effect upon the long-

term attitudes and concerns which residents have concerning the hurricane

hazard, nor are they associated with anticipated behavior in the face ofI

an imminent hurricane threat.

U ~~~~The results noted within the Lower Florida Keys are generally con-

sistent with those observed by Palm (1981b) among buyers in both Berkeley

and Contra Costa County, California. In both studies the survey popula-

* ~~tions were wealthier and better-educated than the national norm and

included no large minority groups. The home search took approximately

U ~~the same length of time in both areas. The home buyers within the

* ~~special studies zones in California (near mapped earthquake fault lines)

felt that "price, investment potential or resale value, number of bed-

I ~ ~rooms (size), and view to be of primary importance" in influencing their
purchase decisions (Palm, 1981b, p. 48). Within the Lower Florida Keys,

I ~~while both price and investment potential or resale value were "very

important" to approximately half the respondents, the nearness to the

shore or canals and access to deep-water boating were even more

* ~~~important.

Some differences did exist in the way residents received their

I ~~hazard disclosures, with the use of contract addenda far more common in

California (Palm, 1981b, pp. 54-55), and Florida buyers more often told

while homes were shown--possibly as an explanation of the increasingly

prevalent stilt architecture. Nevertheless, disclosures had minimal

discernible consequences in either study area. In the California study,

I ~~~22% of those respondents receiving disclosures felt that the information

made a difference in their home selection, compared with 12% of the

Florida Keys home seekers. The slight difference may be explained by the

finding that the primary home selection motivations in the Keys inevitab-

ly bring home buyers into the hurricane flood zone, but in California no

I ~~~environmental amenities are attached to the special studies zones.

In both studies, however, location within a designated hazard zone

is expected by new-home owners to have little affect upon the price of

their home or their ability to sell it sometime in the future. Indeed,

Palm (1981b, p. 54) reports that 88.4% felt "that location within theI

[California] special studies zones would make no difference," while 72.1%

of the Keys house purchasers stated that flood zone location would have

no effect, with an additional 15.3% expressing uncertainty as to whether

or not it would have an effect--findings which are quite similar. The

purchase of hazard insurance (flood in Florida and earthquake inI

California) was positively influenced by the distribution of hazard

information.

The relatively low proportion of home buyers who indicated that the

hazards information influenced their home selection should be considered

in the proper perspective. Indeed, considering all the informationI

provided to home seekers, Hempel (1969, p. 56) discovered in his

Connecticut study that "only 17% of the households reported the broker

[real estate agents or realtors] had some influence on their choice of

homes." While realtors may influence home buyers, Barresi (1968, p. 63)

reports that the majority of ".. .agents attempt to direct the prospective

buyers to particular houses and areas based on their interpretations of

his desires, needs, and financial ability." Furthermore, the real estate

agent is able to exert subtle control over the buyers by the selection of

homes or areas which are shown, or even in the selection of clients.

Newcomers to the communities are typically most reliant upon real-

tors as information sources (Palm, 1976, p. 269). When home seekers have

preconceived notions concerning residential areas, information provided

by realtors may have minimal effects.

I ~~~~~~~~~~~~~~~~~~~~~~~~~37
Buyer acceptance of the brokers's advice is likely to be signifi-
cantly affected by the consistency of this information with the
I ~ ~~~buyer's preconceptions and existing beliefs. Thus, the broker who
rushes to correct the "misinformed" buyer may create suspicion and
confusion which severely inhibits his subsequent efforts to help the
buyer (Hempel, 1969, pp. 39,41).

Newcomers within coastal areas often have preconceived notions about

I ~~coastal hazards. Windham and associates (1977, p. 32) note that many

persons had already decided upon evacuation behavior before they moved to

the shore. It should be further noted that many purchasers of coastal

real estate are not exactly newcomers. Within the Lower Florida Keys

over a third of the post-1980 home buyers had already lived in the area

I ~~for at least a year before purchasing their present homes. Furthermore,

since many of the coastal properties are purchased as nonpermanent resi-

dences, their acquisition often "may follow a period during which yearly

or more frequent vacation or tourist trips are undertaken" (Gober and

Mings, 1984, p. 172). Indeed, home buyers decided where to migrate

I ~~before they decided to move and select a home--a characteristic of the

migration of retired persons. Thus, it is not surprising that realtors

may have little acknowledged impact upon their customers' decisions.

* ~~~~The findings within the Lower Florida Keys study support the

California results that the disclosure of hazard information to home

I ~~buyers had only minimal effects. Not only do most real estate agents

disclose the information at the time and in the manner least likely to

cost them a sale, but the recipient is often not inclined to act upon the

information even if it is given. Indeed, the majority do not even want

to receive such information. Some buyers, upon hearing the warnings, are

* ~~~likely to assuage their fear by claiming that they are less vulnerable

than others, an observation of Sheaffer (1960, p. 73) which was also

1 ~~noticed in the Keys. As Marston (1984, p. 2) notes, others "informed of

the potential risk, still purchase the house because she/he plans to live

in the house only 4 short time...." Where retirees dominate the popula-

tion, as in the Keys, such a conclusion would not be surprising.

An implication of these findings might be that hazard disclosures

are of little value except to ensure compliance with legal requirements

or to eliminate possible liability claims against the seller. However,

even influencing one in five buyers is better than nothing, and when the

proper methods and timing are used, the number reached might be higher.

Although disclosures can have a positive influence upon the acquisition

of flood insurance, neither disclosures nor the National Flood Insurance

Program have done much to slow the rapid growth of coastal population.

This population is not unaware of the possibility of future hurricane

destruction, but our research only strengthens the argument of Saarinen

(1982, p. 1) that "...there is increasing skepticism that more informa-

tion by itself will lead to appropriate adjustments."

REFERENCES

Baker, E. J. and J. G. McPhee
1975 Land Use Management and Regulation in Hazardous Areas: A
Research Assessment. Program on Technology, Environment and
Man Monograph #8. Boulder: University of Colorado, Institute
of Behavioral Science.

Barresi, Charles M.
1968 "The Role of the Real Estate Agent in Residential Location."
Sociological Focus 1, pp. 59-71.

Burton, Ian and Robert W. Kates
1964 "The Floodplain and the Seashore: A Comparative Analysis of
Hazard-Zone Occupance." Geographical Review 54, pp. 366-
385.

Field Associates, Ralph M., Inc.
1983 Preparing for Hurricanes and Coastal Flooding: A Handbook for
Local Officials. Washington: Federal Emergency Management
Agency.

I ~~~Florida Coastal Coordinating Council 3
1974 Florida Keys Coastal Zone Management Study. Tallahassee:
Florida Department of Natural Resources.

Gober, Patricia, and Robert C. Mings
1984 "A Geography of Nonpermanent Residence in the U.S."
I ~ ~~~~~Professional Geographer 36, pp. 164-173.
Hempel, Donald J.
1969 The Role of the Real Estate Broker in the Home Buying
Process. Real Estate Reports #7. Storrs: University of
Connecticut Center for Real Estate and Urban Economic
* ~~~~~~Studies.

Marston, Sallie A.
1984 A Political Economy Approach to Hazards: A Case Study of
California Lenders and the Earthquake Hazard. Natural
Hazard Research Working Paper #49. Boulder: University of
Colorado, Institute of Behavioral Science.

I ~~Mitchell, James. K.
1974 Community Response to Coastal Erosion: Individual and
Collective Adjustments to the Hazard on the Atlntic-Shore.
I ~ ~~~~Department of Geography Research Paper #156. Chicago:
University of Chicago.
Monroe County, Florida
1977 Ordinance No. 8-1977. Monroe County Board of County
Commissioners, Key West.

I ~~~~1983 Ordinance No. 30-1983. Monroe County Board of County
Commissioners, Key West.

Palm, Risa
1976 "Real Estate Agents and Geographical Information."
Geographical Review 66: pp. 266-280.
I ~~~~1981a "Public Response to Earthquake Hazard Information." Annals
of the Association of American Geographers 71, pp. 389-399.
I ~~~~1981b Real Estate Agents and Special Studies Zones Disclosure: The
Response of California Home Buyers to Earthquake Hazards
Information. Program on Technology, Environment and Man
I ~ ~~~~Monograph #32. Boulder: University of Colorado, Institute of
Behavioral Science.
Platt, Rutherford H.
I ~ ~~~1983 "Local Government Liability Regarding Coastal Hazards." In
Preventing Coastal Flood Disasters: The Role of the States
and Federal Response: Proceedings of a Symposium, ~Jacquelyn
I ~ ~~~~Monday, ed. Natural Hazards Research and Applications
Information Center Special Publication #7. Boulder:
University of Colorado, Institute of Behavioral Science.

Saarinen, Thomas F.
1982 "The Relation of Hazard Awareness to the Adoption of
Mitigation Measures." In Perspectives on Increasing Hazard
Awareness, T. F. Saarinen, ed. Program on Environment and
Behavior Monograph #35. Boulder: University of Colorado,
Institute of Behavioral Science.

Sheaffer, J. R.
1960 Flood Proofing: An Element in a Flood Damage Reduction
Program. Department of Geography Research Paper #65.
Chicago: University of Chicago.

Simpson, Robert H. and Herbert Riehl
1981 The Hurricane and Its Impact. Baton Rouge: Louisiana State
University Press.

Slosson, James E. and Gay W. Havens
1984 "Legal Liability: An Incentive for Mitigation." Paper (P/P-
16) presented at Natural Hazards Research and Applications
Information Center Workshop, Boulder, Colorado, July 15-18,
1984.

Wilson, John D., Daniel L. Trescott, DeeEll Fifield, and Vera McIntyre
1980 Hurricane Hazard Mitigation at the Local Government Level:
The Roles of the Building Code and Other Development
Management Strategies. Tallahassee: Florida Department of
Community Affairs Bureau of Disaster Preparedness.

Windham, G. 0., E. I. Posey, P. J. Ross, and B. G. Spencer
1977 Reactions to Storm Threat During Hurricane Eloise. Social
Science Center Report #51. State College: Mississippi State
University.

NATURAL HAZARD RESEARCH WORKING PAPER SERIES
Institute of Behavioral Science #6, Campus Box 482.
University of Colorado, Boulder, Colorado 80309

The Natural Hazard Research Working Papers series is a timely method
to present research in progress in the field of human adjustments to
natural hazards. It is intended that these papers be used as working
documents by the group of scholars directly involved in hazard research,
and as information papers by a larger circle of interested persons.

Single copies of working papers cost $4.50 per copy. It is also pos-
sible to subscribe to the working paper series; subscription entitles the
subscriber to receive each new working paper as it comes off the press at
the special discount rate of $3.00 per copy. When a new working paper is
sent to a subscriber it is accompanied by a bill for that volume.

I The Human Ecology of Extreme Geophysical Events, Ian Burton, Robert W.
Kates, and Gilbert F. White, 1968, 37 pp.

2 Annotated Bibliography on Snow and Ice Problems, E. C. Relph and S. B.
Goodwillie, 1968, 16 pp.

3 Water Quality and the Hazard to Health: Placarding Public Beaches,
J. M. Hewings, 1968, 74 pp.

4 A Selected Bibliography of Coastal Erosion, Protection and Related
Human Activity in North America and the British Isles, J. K. Mitchell,
1968, 70 pp.

5 Differential Response to Stress in Natural and Social Environments:
An Application of a Modified Rosenzweig Picture-Frustration Test, Mary
Barker and Ian Burton, 1969, 22 pp.

6 Avoidance-Response to the Risk Environment, Stephen Golant and Ian
Burton, 1969, 33 pp.

7 The Meaning of a Hazard--Application of the Semantic Differential,
Stephen Golant and Ian Burton, 1969, 40 pp.

8 Probabilistic Approaches to Discrete Natural Events: A Review and
Theoretical Discussion, Kenneth Hewitt, 1969, 40 pp.

9 Human Behavior Before the Disaster: A Selected Annotated Bibliography,
Stephen Golant, 1969, 16 pp.

10 Losses from Natural Hazards, Clifford S. Russell, (reprinted in Land
Economics), 1969, 27 pp.

11 A Pilot Survey of Global Natural Disasters of the Past Twenty Years,
Research carried out and maps compiled by Lesley Sheehan, Paper pre-
pared by Kenneth Hewitt, 1969, 18 pp.

12 Technical Services for the Urban Floodplain Property Manager:
Organization of the Design Problem, Kenneth Cypra and George Peterson,
1969, 25 pp.

13 Perception and Awareness of Air Pollution in Toronto, Andris Auliciems
and Ian Burton, 1970, 33 pp.

14 Natural Hazard in Human Ecological Perspective: Hypotheses and
Models, Robert W. Kates (reprinted in Economic Geography, July 1971),
1970, 33 pp.

15 Some Theoretical Aspects of Attitudes and Perception, Myra Schiff
(reprinted in Perceptions and Attitudes in Resources Management, W. R.
D. Sewell and Ian Burton, eds.), 1970, 22 pp.

16 Suggestions for Comparative Field Observations on Natural Hazards,
Revised Edition, October 20, 1970, 31 pp.

17 Economic Analysis of Natural Hazards: A Preliminary Study of Adjust-
ment to Earthquakes and Their Costs, Tapan Mukerjee, 1971, 37 pp.

18 Human Adjustment to Cyclone Hazards: A Case Study of Char Jabbar, M.
Aminul Islam, 1971, 60 pp.

19 Human Adjustment to Agricultural Drought in Tanzania: Pilot Investi-
gations, L. Berry, T. Hankins, R. W. Kates, L. Maki, and P. Porter,
1971, 69 pp.

20 The New Zealand Earthquake and War Damage Commission--A Study of a
National Natural Hazard Insurance Scheme, Timothy O'Riordan, 1971,
44 pp.

21 Notes on Insurance Against Loss from Natural Hazards, Christopher K.
Vaughan, 1971, 51 pp.

22 Annotated Bibliography on Natural Hazards, Anita Cochran, 1972, 90 pp.

23 Human Impact of the Managua Earthquake Disaster, R. W. Kates, J. E.
Haas, D. J. Amaral, R. A. Olson, R. Ramos, and R. Olson, 1973, 51 pp.

24 Drought Compensation Payments in Israel, Dan Yarden, 1973, 25 pp.

25 Social Science Perspectives on the Coming San Francisco Earthquake--
Economic Impact, Prediction, and Construction, H. Cochrane, J. E.
Haas, M. Bowden and R. Kates, 1974, 81 pp.

26 Global Trends in Natural Disasters, 1947-1973, Judith Dworkin, 1974,
16 pp.

27 The Consequences of Large-Scale Evacuation Following Disaster: The
Darwin, Australia Cyclone Disaster of December 25, 1974, J. E. Haas,
H. C. Cochrane, and D. G. Eddy, 1976, 67 pp.

28 Toward an Evaluation of Policy Alternatives Governing Hazard-Zone Land
Uses, E. J. Baker, 1976, 73 pp.

29 Flood Insurance and Community Planning, N. Baumann and R. Emmer, 1976,
83 pp.

30 An Overview of Drought in Kenya: Natural Hazards Research Paradigm,
B. Wisner, 1976, 74 pp.

31 Warning for Flash Floods in Boulder, Colorado, Thomas E. Downing,
I1977, 80 pp.

32 What People Did During the Big Thompson Flood, Eve C. Gruntfest, 1977,
62 pp.

33 Natural Hazard Response and Planning in Tropical Queensland, John
Oliver, 1978, 63 pp.

34 Human Response to Hurricanes in Texas--Two Studies, Sally Davenport,
1978, 55 pp.

35 Hazard Mitigation Behavior of Urban Flood Plain Residents, Marvin
Waterstone, 1978, 60 pp.

36 Locus of Control, Repression-Sensitization and Perception of Earth-
quake Hazard, Paul Simpson-Housley, 1978, 45 pp.

37 Vulnerability to a Natural Hazard: Geomorphic, Technological, and
Social Change at Chiswell, Dorset, James Lewis, 1979, 39 pp.

38 Archeological Studies of Disaster: Their Range and Value, Payson D.
Sheets, 1980, 35 pp.

39 Effects of a Natural Disaster on Local Mortgage Markets: The Pearl
River Flood in Jackson, Mississippi - April 1979, Dan R. Anderson and
Maurice Weinrobe, 1980, 48 pp.

40 Our Usual Landslide: Ubiquitous Hazard and Socioeconomic Causes of
Natural Disaster in Indonesia, Susan E. Jeffery, 1981, 63 pp.

41 Mass Media Operations in a Quick-onset Natural Disaster: Hurricane
David in Dominica, Everett Rogers and Rahul Sood, 1981, 55 pp.

42 Notices, Watches, and Warnings: An Appraisal of the USGS's Warning
System with a Case Study from Kodiak, Alaska, Thomas F. Saarinen and
Harold J. McPherson, 1981, 90 pp.

43 Emergency Response to Mount St. Helens' Eruption: March 20-April 10,
1980. J. H. Sorensen, 1981, 70 pp.

44 Agroclimatic Hazard Perception, Prediction and Risk-Avoidance Strate-
gies in Lesotho. Gene C. Wilken, 1982, 76 pp.

45 Trends and Developments in Global Natural Disasters, 1947 to 1981,
Stephen A. Thompson, 1982, 30 pp.

46 Emergency Planning Implications of Local Governments' Responses to
Mount St. Helens, Jack D. Kartez, 1982, 29 pp.

47 Disseminating Disaster-Related Information to Public and Private
Users, Claire B. Rubin, 1982, 32 pp.

48 The Nino as a Natural Hazard; Its Role in the Development of Cultural
Complexity on the Peruvian Coast, Joseph J. Lischka, 1983, 69 pp.

49 A Political Economy Approach to Hazards: A Case Study of California
Lenders and the Earthquake Threat, Sallie Marston, 1984, 35 pp.

50 Restoration and Recovery Following the Coalinga Earthquake of May,
1983, Steven P. French, Craig A. Ewing, and Mark S. Isaacson, 1984,
30 pp.

51 Emergency Planning: The Case of Diablo Canyon Nuclear Power Plant,
June Belletto de Pujo, 1985, 90 pp.

,DATE DUE

Id~~~~

GAYLOR[ No. 2333 PRrNT�D N U.S.A.

3 6668 00004 7698