. . . naval-Ila“! . 1
V . . . 1 gnflmmvflhwfimmtxe.
. . . .1 ......fiazf .V
.V .V . idaflkmkk. .... .V V .... . V H. . ...
.... Swag: _ ...»... . hafifig ......I V .V . . V- .1. V ...... .-r
. .... fiwn?:. .3 e fifififi. Vur .V . , V. .. . .V V. . V. ,V ... .... .PfVVVVVVthfly. n
V V ..Fv KEEN E.- iaavvn Vonvkuwfi ... . . r V L.“ . . 1 . 1V§5Wf§gfihhuf .1.
V . V y 1 evafl k5 .. 11.: . V . Vst. VV .. A )1. . 3:41.11”: 5 V
. 1. .Ifi. $59.“. .> «M a..." a . . . ~ I. V V. If t
, I.v...t> v; «.V . . V .1: w a
... V.» r Q. -. ’~L&K~Tfi ‘1' uimrith 'q‘ .VQV “MN. 1 .V 1 V 4 1 V’IPI. - I I
. . . a 7.... 3' #5.“ a3 nth .0... 12". V V . . . . . .
1 .V . . . m... Wfififibfiu m ..u... 4% .1.“..an V. V V.
.V .V .. V1V....VV...w:.VV% V
. 6 .r: a. 4: Tu...” . V. V
V. . mfififizakv V... V. w .. a
V ...... 17. ..a v
fiamum: .

.' .at‘kurmsfivvv
2%». Magda.“
1. V1v|.uV.)V.'1V
V_ .V V. .1 V .VMVVRFanmflfiwgflV ..
1 q . V. V V. V1 ... s? .v. .1. .. 1
. V V V V 1.21,... V V V .erVVnmwfiQm ..sz
ruff. 1. . . . . 1 . i1 V.
1 . .. v 1 .N \- VVV‘wV 15km. V . V ... V c.
V1 . . . V3 V1 V 1““. at... V.

w...» rawhihu 3.3%.. .... -1.... a...

  
         

        

       

 
 
 

  
   
 

1
. 11...? 51. n .
V. is.) V 1th .
V .1 V E V V. V .38“. :Q . .
VV thannfiwinnwvfiwuna V. f. V .1. hung... .
5% W 3. V 4mm.) V . {thfnu‘ y m A... .
a» g... \
V 1 fig...) V V .%V\3\s.u
- V V ‘ V V V . . l
...: ru. ...Hohm. ... 31!... w . V .11. ...ka V21 .... V 51%.
u .m x .1....Vmwntaux. V. {.203 V .
«......Jnfifi Vang ... V.V
.V i. m.-
1 if V
. s-

V w
v . . $1 .5 J.“
.1...» L V grant...“ 4
_ . V ...... 3
T, ng .2321. ...... s
\52\\1J.£,u. 1531i§vlfikgsw . ku.‘ .
. . .V V V vsV$v.VV..J.1fiat3x.:ixtf.\fi.fix1shxm. \.V. V
V. V V . ... Lad}... V V V 5.»...fi3aa1mwuw... h
I s9”, 71 t! Y.um1‘fi‘ figural .1 54-771“; .
wan. ....umw 2.... . ...
.3 Wm“ ‘Qézfiflfiflz .b
. . ‘1

V
“v 1.3. m. 4... #5
q . ufi§hfiwmaw 1.19%.. V .1515...
$1.... :flwdfiamaflz V $1...." Q .12?
1.. . Ezzwmmcak$

 

...IL V
VqV . . .p 1. u
z)! .35. 5V (V 0‘ch
V 1.), ..s 2. V V1
V 1 “fl 4 ti»... .
. - . ... 1 .33... gauzmfi V .3. V .V .V
.. 31$" .1.”.h.12....12...V $1.141... 1.131331 Q V
V p .l 1 . H. l I} 1 V V . V V V V
t .. 1). (hm... 3.03. 53 ).Vtw\fla.ww11mufiaflamhm1uwflfix .215)... L1 151%” \d. guru“... V v\\ is:
o . ”I... “.02“... 2. 111‘..$uu. 1.511....) kTanevhu £311.... iflx‘isiwggdfl 1ftzllt 11~113fihfl|13§1v\3 . V 3g!
. Veal». b... . RV... 1 n V33. #1051511: ...«Jmfigfindmmuh: a. V wflfiikfifilx 3.4.33.1.
... V V .1.- ......shumh .yifgaufiwfimgé. Wagfizfi. .2313
1 I . 1 V 1. .1 ...1 . . V1
. V4 VvhaFLR V Vim. 10.55%. 45.1201?” 31....sz
9." a .731»... Immaliflflx.
Hal. .1 .1: :22m . V

    

      

“592‘?
......
\175 gig 2|!

3AM.

  

V 3 y“
hivu‘i‘fia:\
2:. =1.- )ilu‘.‘ 3v? fiflfixhvvaaa~
» 23...» V- v, ..V. 333...? hu. ...
1.11 V leak... 1. ......n i...
.11 \t‘fw i V . V) V S V
3 x. \3‘. “can... 32¢§g¢§nh§d¢§h 11-11%
.) 1 V1.55... I. I: 1 V»
1. a. .7. 12.11 ahfihvzwfihwngéfls
V 3...; -..: 1.1131322“... him»; .uflwghfi, V.
V V V5533. 1.51.1.1? huiviiii... .53..
.V 1 1 ...... V 12%.. a {haw

5.1.... 54.»
.1

V v ‘1 ViAfiLhU.
V P dun»-.. 1m?
2111.. 3%“ V khfir‘sfln‘lufi
. a“?! \vmluwaflnu? V
Au1......Lr4nfiHEMV».....Vnmfinfiwmuwa.§ V V
. .1 1.11m... V.:n.114uhwfl.nuh§flhfl» \ 1P. 1 q
4 . ..V ... 1 V1, . V a \ .
V . .21 xyiahuuwbb... V 33... 1
V .5 31.1. 3 51.1.7!
Ag. V V

         

       
    

 
   

 

V 1.
9 ..
V .15- ‘1‘\#‘»u 2r: )7! .1.-
514 Lil}... v} 1,)! i .
17 ‘11.... TIL: \\\. JI!\3«.\...5
3)): 3h“ ......»H a «3 $3113..
...w. .1: a .V 1.P3.Lunm.$+mn .mmmmhvmfia ......wmaua 3311b.fl>.m.
.1 .1V . . ‘ V V1 .1 1 u. 1
...... 3C 3.11 1).. MN: :17 ...-J11. 1313 . . $51.15.}...
... V ....w s» H§VuvLwa..1m1u? £38.19... 3119.. 51.91%». V 31%.? .5... 11.13.141.135.
6 i V V .3‘ u‘. {.11 ..{1\ 1.1... 1 41V .111‘ . . .1.. V
1%111571 V 311‘}an 7... 335331 33.1.1518 ... in! 3111,11,!) S :- Adiizfipua 51;}.{1
V V 3(1):...3L1»).1£55714(.1V1\4n155.1\2~rwu§v1521.13x11.\ #IS‘sxsaISIIaagoazxfi . Ariss‘svsifsithé
V 1izu\\buv.131. 33: 2.33%.. 112:}..131’11111 \T.J.131‘2)|\ 533139.21). 5.1. 51‘ lg 1h. Vl{33‘3)\ztxi
15.12.)? .. a}. .15.. .. axwfizfizfiuuuo J...\...h¢..~§tus}v+.{axkxliayrsxzvafirazww‘111y I; V . 1.15.... fihfiuvfiniuaflfiii.
.1!’mu11\\1‘1$0£¢l4.41 . 1115\kw. .715! 8.3::‘3153 5 15V $313111?!» *31711 5Y1i11i‘1tfx «€511,731 $112. 1) avaiii}: 72 1.113).}; all")? I! f!
31‘ if 3V Anna 3 . s: .. 1.. . T \I 31‘ 1 ...:x x. of 114173. 33.61;»)!!! 1. 1.4V .. . 1V 1V 4 I 311311
£17. 1 L. 5.11.31!sz1%;1?hm\cfi§3hfihfi}iumkflmVisitmxag‘nyfixtagfihvflfifibnfl \ wsgfia’zrfiififlu‘sh‘15 H.053». Nuggvihrqnfihvufinglkuuufitzg “:33...”
3.... 1 VV 1!; \. 15$“ 15 2-1.3. 7 I \ |5 I“. .. \111‘611 1‘15 3 153.111.111.31“! .1 8255‘ 3.. V ‘53!) $11.11 \1 V .
...?!«I. 1. v? 0* £111; .2)» #341,).{51 1.151311125313111 )1 $1.1z§v\7s32.$1.1.1$.i¢3\l\ ...}. ‘3179‘QHiw‘13LNW113‘113 Vlad)». ‘31): i
; v If... ..r? 11. 1. 71.... 1.31... 1). gt)... r 1.3.1.1171 11 V .3. ...i $1.. .. k” V v! .11}. 5.. )3
VVInnazhfiil \111Wwduu. V .53. fivxxinwwudsafi4....1..1Vfi...JJwa§..nVVV ywbifisijn 9H1111§11§¢13$sr washmfikmwflwauyflhzilfif V a; .1vgamy1mwdua n
5 17th.??? . 4 11 3a 3. 1V\ 11.1.5115: 33‘, \\ ‘11}..7...2k\ ...: 3.344....1V 3.1 I)... “3%); a? (1.5)}.53. s . a V1 .V .
1 1‘ \Vnilitzfinc. hofiT15\.. {tiliflini‘ 4.311113! T155 \(37 ....)11‘)‘ .\VIJI 5.1.! izil’rwwnw 151.11.! k ‘3‘: .3
. I... .VV AWE} ...Pxxsfxszxu. 1.1.1.13... \fixvialtfusfiaual.‘ xv..1k).....r.i2. arm... I; an. 7%....3 .L ...
a 19 1 3.1 ......7».a..y\..1}4r¢.s1..3>nsass .ia?1?1i~i7i..i4{!sixaxfitfis§1 €11“ $.13...) fil\§r7~lfih. 71 s .1.... afluwzl. bi.
in»??? $3. 3.3.)?!{55 341.11.11133311173‘:111,.111w(:.n\34sw\73 111‘) 3: {e111 I? V .3 AH]; I)! 5 1 l:
a. V :11 (Jig; ».v)15k]1177\i J13§4fi912x~iSWI)11..\.1\(1)1I1§g13x7$1i1331fi .71. 9..st ~.i.|1\~t1.GV\N1. D). Vsi‘n... V .rrf
as}:zzx\x\21.51w?»§13§~¢x¥u%sw\i . 13545.5... s‘\.1:}.flwf.1vaixs$rli 3:33.}\ I I «vyiV .4112.
. 1.5.1.. 71355.1)... i5: 85)}. 2.75 151.... 3,1134%.»113119;.Vfi.nu\.x\v‘sf5\\fl1i\vy}bl v.0... \. \3. EM. .2:
‘ \\}.nhw..sfl.\§\u§«h.z~hhi)xrwbfiw 12.1.» \1):\..Iv11 8.. .\711?I!1§jl\.§xvs«€ . 5h: .VV. $4.14. 1
«(1117 3).. 1% ......» 3.341.223.Dw‘Jhquiifii1-5Y131hrlfl wfi!1..~.vf7113\)§‘ 73‘171\¢uu111o3 .VV. 5». fl. .1.. 5!
{$135.1 1}».nisixt1ii )vaiTV;.1\d.11..~.J 11.[.Vfiwv§:\!111v.l¢fi\!.2l\nixln 44.1%....
1131.311} .171. 111 V ... «£1.31... x $123.??? I! 4731)..) 2.)... .1... Inna}...
Dalfwlv1‘a‘1hu.\ 53v flyauq...Hannah’s..211331Hhvfipdu’oifaawi3.5.112? v
£§R§5¢fiii~ 1.51%20.13.11.511assurififioagiiizdéufl$13.31....)
V , 1 V V . .. 3‘ ... . VV 1» 5 l 3
fiztuuflimsjwl . 1:31.32 17552.11v\1311)fl13{?2£533. 1.9.1le
is ’l y-‘l .\ Aul.CLI"‘A \“. 1‘.“a‘1\‘ in 3.4.1 1‘
V 1143x119. ...nfi..§..s!P¥.. 131.112.13.31. 1
11137110 551 .V . . .
fig. \ . V

         

I |&
. 2 11:... iffislhhfi
1511111”? .) .runur.tl§t4\ .
I'AI-V 1‘3. ‘ \I "I’
V L a 24.. It‘d 1-H...- gxiflahvx? Psrffihnfiia .. .51.. #3.”. 1
L1 1 .o uifl V I n \Nlu V 1
.V 13153.)». 11).;1‘31231.’ 3...}. g. N. ‘ _. a . 7...} .
. {1.13.1.1 it). V3551P>jz1fl7351li~ .... {\«LHV ~ 2.3).. in 1!,
3.11.5 7. it rV U‘ .. hi. ‘1. 3.. “55137:...
\SSqu1fi2$.w,.m11§hv?.¢b.m«.mhfinn§hfi.1fl.fluttfiumwdflhniz1.-.. V ..
. buflfihfivvxfilknur...W’HfiwwmiflmunfilhflvtpflxV V1waiikJHT. 3.571 .V
.VV ....th yzflfii}! 5.53.1. .....lfixflnhb V3313)! V . V .V
.. F351,! .131Vl.\1u.~1\~1.}1v1\111
.V 415?? :71...

.
12.3% 313?: 14.1%.”?
5.11! V

     

gig... V
31v} 5!!!! . ..11131 .V
$1.} 1.3)! yr): 551 \s.
11111:: . 2.251)... 33.1.. .052. I.
wnzarska . {éfivts

‘ 1315131 20122.}.

‘3‘. 1!.nssxi31csg I? 17!

V . .1.. .VV. hkiaku 3.3.. *1

1€xasmfiwu§zufiu§15 ......V .

. 151$)...‘3‘1Vv M. ...

7131:3111; 41.1114“?

9.13213...

                                  

1-.....
V .... 1
V V. V .. agBéhhq.
11: .33. 11192121179. 141133431141}
7 i..3n.24...h.¢3évmmm.s .fivlfilfi.Vyvafiflxéaifivfshfld .L
I!“ u 51.)...er 533251)) T! 114?»)!!sz 35.1
. . f‘awuq..51 J!» 1.113.715.1415 ...)rnviifiab.
V V gfiaairfigrhqbwfi? ...4331311; 21.1.52... .4
.1.?3111 \ziriihs;
'341d‘flfléfiflifififlwuum5T 1Wquj57111§ {Muiahmf Val.
.. In... J...?1e.121. aifi 139.4
3.. V .VV V, ifiiu~131~f11§§3V€l
I 56420314 H! .h. W. .1131)
‘1‘113; xi! ‘0 . V ’1) u-
}. 31. 3141.: 1.111 .3181. V V9...
2. 3
1.1.3...

.
.V .V .V.1

1 ..JVWanrcyhfltunfiuhyfzu
: 34.31.13“: Vzafisifiztzzdwni 1!...

35.30" .3 . .1 1 1 1 .
V siwgwflmfifii . 5‘ ‘11:? . 15.“.
v. .. I ((13? {$.& 1.x h... . 13.111311
uwflwgdwbzfiqkwwl‘vfi 1i1<psxx.ffl$>:1 gxaflu41 a. ‘55.. .51th “M11135”.
u\-....~¢14HM.I{2¥vlu1.VVVV1.U¥)...V2.V1....»yasffifiax321 finglenaflhfiwmfidmmedfin; 3:13. q... T... V.
1513\51.‘ v. I . V . Vi \ V 1
ggwal‘nfiahgxrsgg.afih:?:suhwu$mww.ngfiflflfid .«ulfWQfifiijlaJHsaz.1flm. Luann: ...V 123...: V
ifg1lihh§xh £1N$l;flluhfit%§\$hiwifiifil “IN; #1135. 3
T1141..\!u§i:§131\53\\fs 2...}. will . V w x. 3:; .
. Sgiijgg“ BQSJfiflvfi11Wng 7+.

5%ng 1%: 3.11.
a 11.; 11h...

       

  

... ....)51
V. ..VaLl
{aha}. .....wnmh?! .1..
f~0f$ . . v
. $2.333? V3. .3 V.
V ... .
1 V ‘v M? §g§§JM¥£W§JV V‘
x 51.. $1 . .. 3.2.17. 11...Pd.2({‘tflss§.x.sv$
.1 «(16512... «.nngfayzt.2£ \1. 1.9...
7?! .I1WnJrIJie).a1-f7»r\1.tohfh .~J .w.
it. 1.13.11”.

_ 2-15
V 1“ A
1% JUL. 1
\V v... u. IJ.N1.1. I" . v
1 a I'lu“
. ‘1 1 . ‘i‘ss.
.z1fi.¥.11£1}3nfi$}.m 311:3....¥;z;V.1..§..1 .0..." .
. Just!!! ‘3‘3sfiafil‘7V 1 1 f“? $8M... EI!.1.1§5MI
LOW-fluid, if“. 29,11311n1ifi 1 mrnhuna‘\.¢i¢d . Vv1vanN1va1Yl
1 :5 xihevgzuijhfii.Y§.v}..\.n&2¢1ffiw V.~_V . \131‘XV. . . gx‘\ . \
. ..l. 11.. . 13.1))... I! V gnu)???
V I 111... 37.173)!!~.1 315 5‘. all. .Jrfikfi‘. 39
3‘ . 1.. V gfiJ‘mmbeix-fiflzfiflnix fifiuwfi. 3.x T1??a~.2.\£. A“
l ‘31. 31f. )9. .... 11.1 2122.8 4|)\ \ €131.31 $71
. g V 7\\§;H\Wfiazshflw1wmu§iiwifiiufl%l. 11 35331V\ 50.»:
g... 0. Vu§iiJLilutwliz
.13.. 11

.
flivfifl§§ ...; .

                

       

      

..
V .1535 135% Hm}.
1.: iiiflrifaaq ~11$i§§ v." 1‘ Etna...
gash-l ’U‘l\\§ .. V
i n11sétzfiixsz-h55siifl ‘Yulixulii V 2139 A 53. huh". V
5‘ 20A! 1" . .
mil... 7 .5155??? 1.11:7; .11... .1.? .53.. 1 . {I} it aux—«lulu:
211...”. v .. .VVxthuuNN-«DVV V11...) .71.. :25.N.41VJ¢R.2A H:Hufln§aui\ 5‘
. the. 5.. 51‘. J\\’1\ \315373511x : .171 V V.
V ..Q!Z..f..>xfivx 1 2......) 15!}‘3‘thwrfigafiaag >123...
V513 53911338 xv (tfimxir qsziiivxkvfix ...-5931 2.))”:
311,334». Vssggtlfq {1.131. 36V. V4211} .fififlgzbgéénxfiv3xt
3‘375ggln.) sldfiua‘ 55.3.7: K1131
V 14.11.! \ 77' it! 13*

V a

u. gill?! fig:

1 . .jfl. {1.16mi-
115.1133 \flniflmldi? .1.
sail. 5.43.? 5
1511137515.»... .
.1 35‘\1‘V V
)1)... ..."! :13 1...:
V 1.171... 1 .Vhi...l..r). ... .
V .. .

1 V ...; 1
Sigi‘i$i¥§
)7. 1.... 1731.7. ..1 7.:
giwuffiffivshh . $1531.21. 1.3“,”
113.1“ ifirzaHLNH‘1v1‘5Mnfi gezgifi (:1

gg’t‘flnz 325”. 2.1-. ‘10“!
0.131.114... ‘ ‘1: again». n. 1
§?.$!;E.zu” 2. u...
5

V. \xJ‘TWi
. zgsiv‘xuwwl 1“;
. .. #:1517195 V‘)...‘ '(i).\.1i§ 1.1.
aguusanfixssair101.69.112.2563uu}
. $531.31; 5?!!! 351’7fi7fylllivi
. 213.113.31.241.“ka 21.?3vnvwwhz. n. £225.92... V
. £7; 15:. 3xr1tv11 ii...NVV)$V.V1hHuJfih.xx
(7.: I533 52...... 7!.15‘3\.)5§I2l.1 V11;
tang...- A‘...vll.44§?3i‘hfl~¢r‘v.l 2.7:: ..1 .1155"; 3315.. 11,15 11.“.
71.1. . . . . is 5.111 11 it. 1‘. want»?
.1.. V V \ $3111igxavnufi rifiqvlii
{1‘ 11‘ xiii-55.4.5313.
Inigx...‘

                

gall-$01. ‘)
.fl3}31\?~(fi§lfl{
«a. Y‘QVgg. $31.01.:IQVhrquinlxiiju‘u. ....vNVYlaq
3 g 1“ ~53; gi‘! ‘(vk-‘Iflg‘t‘ 31!
....- .. ... .v 5} $113112; 1 \O. I 1.41.1138 ’73?
xh. kfihuuWIh'auW1533VN151ukfifl7thil. 441.58 a‘
.. «gigg‘tfl‘fzhh'flighi1? u!“
22n|

. fl 3»
V V zflxufifiwhugunux
1 . xngdzuflfiiar: V

        

1 ‘ 1141.25}
. 72 4.3.1:!!!
at. . fl 7 “It:
tr V . \jxfisqxflhfaéz!£ V
$75 71\\:~H.u\.w,gsfd${7\3\51§1 li'V—wé.111
‘54:..nrli31iaififiluuflh rx ‘1‘...I.Nfi171..1 ...
.1711? u. 3:. it...) 3 z ‘1-
1411112: 53v\\\..:4li1¢$1 V?“ y. {tn-.641.“ \ 151..
.l. 7.. 1%.»? 243.13.:
.5 a $3... afigx‘jax.§1v}1\{1} #v 3.30....)
1... V :1;
31A 1&4

. V . 51...!
q! ...!
It...‘ 13):: . .1.. . V. 3.1
31‘11’!..\7.§\..V..§.3r... zsxiizitfiv
V ‘4“! finfii?§$§ws:~ui\
\'.11. 3113551111 ..
’1‘7LL11!

. u.
l. 1.3:}? ft
V.‘ thLfflt:
V It. 33181 V
V a 1%1afiu9311 .9121.
.. 311.011....1 Ygfi1krfita w.
3 ifltfiz. {1231s. (14%
.. V flunk. $111!;q

    

   

   
  

  

5‘s
Y? ‘5 1
{in}! 1!... ....vlc: “VV1.. V. ... .. .51.:m1“..1. I .

1§.$I§.V\.n $§Vl§§ 3.011.931.“

1 9 3317!..‘11 ‘\i~ufl§.fl~’§3§§§d

. lg. n 111..V\u2”fi!.5.1~l..._fsh uh . . . 1¥1t lgfhuik‘aflu
1 V 1119\1Lflu.1111§1.i1 2.715395. if 75"}
. .11“. 31:5}Ja‘iiis‘1f‘2;
V ‘fhfluuru1155‘511‘fii 131.113.1119
$12§}fl1:f,.ia!$111\11~k1hl41
\ ’31“... i. ‘1‘?" 51$“ ‘37:...
ii.- 1 it};
sit

      

   

1.131.:
1142 . 7.
3.533124%... Wyn.
{32.434221}.
125915.;13‘1
. 3-2.1:}.
. 1‘4
. w: ..ila....21%+¢. 1......
F1, v1.2. 31132 flrflnduufinéu
3129.-
£55.“... 5: 3%?

V 1 fgtzx. - 21......

v , v .

21; g.

1b.; .V

     

       

A... I .115. ‘.
S‘Iisnlrgikfln‘ 3391
521 9.11))? it“. t: ... ..
2.1.. I 355533....3”? H2“ 113...)..191? twhuuxfiu.
. 1.3.113. inhfé ......ré . ‘11:“.ku $2.144. $1114. ... . . .
i1tx11‘13:}§~vflu8\1kwfltfi fl w.:!.l~....fifl~-r« V» . .... V 1 . . r . V 1‘ .
$..\5hv1flt5111\1{51k2‘ 3.24;“: v. . “1.3....“ .....m\._ V 1 ..a.§\ 1 _ .1 1
s s! M1081. ofigfloivbf 1.1.1.1321 ... &. ... . 1.x. 5E1
V t 1 .... .V: .VVV.v.VV.V. .1. ... 1 1\
‘ ......V...,JAJA.V:C.=V_.5::.1..V..V. o x. V ...
)Qr. ‘00.? u . V 1 1 . CE»
.17?! I » ~ .h
:1... VV..a«1:3:1.~.. VVuV V
1-1.~ TV»..-
. rill-HP.
..“ngna... . ...

    

- t -: .-'- ‘. -:' I. . ’% 13“.}; .1. -...':.‘. =
- ""fl ' .1.." ' '
1 . ' ) 11diilfi1d0mfifiythnt£hg
thesisentitled -

ESTIMATION OF THE ECONOMIC 'VALUE OF WETLANDS:
METHODOLOGY AND APPLICATION TO SOUTHEASTERN MICHIGAN”

presented by

Chaéles William Abdglja,

has been accepted towards fulfillment -
of the requirements for

M.S. dPgrmpin Agricultural Economics

 

Major professor

Date FWD-*7 zztfiaz’

0-7539

 

 

 

RETURNING.MATERIALS:
LVIESI.) Place in booE drop to

remove this checkout from
w your record. FINES will
‘ be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

,. , A ‘ ‘ w
' / ..o e
I " Wines» Q9"
9“?me I . _ . i
I‘— m». ‘
..., 850‘
\I‘ // FA 0 84 3 ‘
L It“; A 51' MAY112005
“553“- ?fl} . .t ( 4
v'fgqunfil a”, ’* em. .1 .,
. . ,, Iil-Ellilééyjjggjl U l, L , ‘
9 0 9n \. ; .2

 

 

 

     

ESTIMATION OF THE ECONOMIC VALUE OF WETLANDS:
METHODOLOGY AND APPLICATION TO SOUTHEASTERN MICHIGAN

By

Charles William Abdalla

A THESIS

‘ Submitted to
Michigan State University
in partial fulfillment of the requirements
‘ for the degree of
MASTER OF SCIENCE
Department of Agricultural Economics

1 1982

    
 

7 ' mm Maximum
MA m. WJWETJM

 

ABSTRACT
ESTIMATION OF THE ECONOMIC VALUE OF WETLANDS:
METHODOLOGY AND APPLICATION TO SOUTHEASTERN MICHIGAN
By
Charles William Abdalla

Greater understanding and awareness of wetland functions has changed
people's values toward wetlands. A framework encompassing political-
economic dimensions was developed here to analyze this change and the
evolution of wetland protection laws. Michigan legislation requires a
permit for wetland alteration. Permit decisions are based on the bal-
ancing of benefits and costs of proposed wetland uses. Economic concepts
are utilized to structure elements of decisions to assist and improve
policy choices. Benefit and cost estimation is attempted. A method-
ology is developed and applied to estimate the value of wetlands in
waterfront residential use. Results indicate that location is an import-
ant factor influencing development value. Value estimates are site—
specific and not generalizable. A review of the wetland-water quality
relationship revealed sparse quantitative documentation making valuation
premature at this time. Estimation of development and natural wetland
values is a complicated task requiring careful economic analysis and

much reliable informational input.

 

 

F..'f-‘..“;"-'-. "5"” l '5‘";

ACKNOWLEDGEMENTS

I would like to thank several people for their contributions to
this research project and my graduate education. My major professor
and thesis supervisor, Dr. Larry Libby, deserves a special word of
thanks. His guidance and support throughout the development and com—
pletion of this project made it a rewarding experience. I would also
like to thank the other members of my thesis committee, Dr. Lester V.
Manderscheid, for his assistance and advice at various stages of my
program, and Dr. Daniel R. Talhelm, for his useful input.

I wish to acknowledge the Institute of Water Research at Michigan
State University and the Michigan Agricultural Experiment Station for
providing financial support for this research.

Finally, I wish to express special thanks to my wife, Katy, for
her help, encouragement, and understanding throughout my masters

program.

1....“- -v‘ '

 

 

"a

wee 51." um: m._ Ell" {M

.i.
.'l'

.ull'

 

TABLE OF CONTENTS

LIST OF TABLES . . . .....................

LIST OF FIGURES .....

CHAPTER I: INTRODUCTION . . . . . . . . . . . . ......
Wetland Resources: An Overview . . . . . . ........

Perceptions and Values of Wetlands . .
‘sWetland Resource Use . . . .
1 Wetland Protection Legislation . . . . . . .....
Economics and Wetland Resources . .
Research Approach . .
Organization of the Thesis
CHAPTER II: A FRAMEWORK FOR THE ANALYSIS OF NATURAL
RESOURCE ISSUES . . . . .
The Political Economy Surrounding Natural Resource Use
What Is the Political Economy? .
Basic Issues in the Political Economy

The Mix of Political and Market Mechanisms for
Preference Articulation . . . . . . .

Interdependence and External Effects

Natural Resources in the Political Economy .
Attributes of Natural Resources . . . . . . .
Scientific Industrialization and Natural Resources

Analysis of Political Economy of Natural Resources: The
Structure-Conduct-Performance Paradigm . . . . .

Introduction .

Structure . .

Conduct .

Performance . .
The Dynamic Nature of the S- C- P Paradigm .
The Role of Research . .

mh-wab—Ii—Ir—I

WNNN

 

iiiv.

 

CHAPTER III: APPLICATION OF FRAMEWORK T0 WETLAND RESOURCES . . .

The Wetland Resources Model . . . . . . . .........
Structure ..... . . . . . .......
Physical Structure ..................

Wetland Production Processes - Hydrological and
Biological Structure . . . . . . . .

.‘\Institutional Structure - Property Rights and Rules
Concerning Wetland Resources . . .....

Conduct . . .
Performance . . . . ..........

Reconsidering the Model with Wetland Transforming
Technology . . . . . . . ..... . . . . . . . .

Structure . . .
Physical Structure . . . . . . . . . . . . . .....

Wetland Production Processes - Hydrological and
Biological Structure . . . . . .....

Institutional Structure
Conduct . . .
Performance . . . . . .
The Model as an Evolving System .
Performance as an Input to Another Period .
Interaction in the Political Economy
The Evolution of New Institutions . . . . . ..
, Wetland Protection Legislation: A Protective Institution

\ The Michigan Wetland Protection Act: A Protective
Institution . . . . . . . . . .

Jurisdictional Boundaries
The Bureaucracy: A New Participant .
CHAPTER IV: BASIC ECONOMIC CONCEPTS FOR EFFECTIVE WETLAND
RESOURCE POLICY . .
Economics: The Study of Choice .
The Concept of Value
Opportunity Cost
Marginality . . . . . . . . . . . . .
The Use of Economic Concepts in Wetland Permit Decisions . .
\ Guidelines from Wetlands Protection Legislation .

Page
20
20
20
21

23

29
32
33

36
36
36

38
38
39
41
42
42
43
43
43

45
45
47

49
49
49
50
51
54
54

 

29m has mam v319q01'l . -‘--un.+1..:"j:;' {mm '.._7i1.--.'

'Eo"1l.'l ’ {'n": I .":-'- .. ' ’31-’-

3

H?
:m

Federal Legislation . .
Legislation in Michigan . . . . . . . . . . .
Informational Requirements of a Wetland Permit Decision

CHAPTER V: ESTIMATION OF DEVELOPMENT VALUE ....... _
Methodological Approach . . . . ...............
What is Development Value? . . . . ............
Willingness to Pay . . ..... . . .........

Development Value Measurement Concepts . .
The Demand Curve and Willingness to Pay . . .
The Impact of Project Size on Benefit Measurement .
Consideration of Substitutes . .

Focus of Development Value Measurement . . ......

Measurement Approach: Land Price Analysis ........
Land Rent Theory and Changes in Property Values . . . .

Land Rent Differentials and the Hedonic Price Method
Tools for Operationalizing the Approach
Theoretical Model of Land Values
Case Study Design and Selection . . .
Statistical Approach . . . . . . . ......
Application of Methodology
Procedure . .
Data Collection . .
Data Modification . . . . . . .
Statistical Analysis and Results . . . . . .....
Interpretation of Results . . . . .
Gross and Net Development Value Estimation
Clay Township . . . .
Harrison Township East Study Area
Harrison Township South Study Area .
Putnam Township
Evaluation of Methodology . .
Validity of Assumptions . . . . . .
Availability of Substitute Sites .
Assumptions of the Hedonic Price Technique

Page

54
55
56

58
58
58
58
59
59
59
6O
63
64
64
66
69
69
7o
71
72
73
73
74
75
79
79
80
81
82
83
84
84
84
85

 

Assumptions in Development Cost Calculations . .
Validity of Empirical Analysis

Case Study Design . . . ..........

Data Sources and Quality . . . . . . ........
Usefulness of Methodology . .........

CHAPTER VI: ESTIMATING NATURAL WETLAND VALUES: THE VALUE OF
FRESHWATER WETLANDS IN IMPROVING WATER QUALITY .
The Relationship Between Wetlands and Water Quality
Wetland Processes Affecting Water Quality .

Plant- Mediated Processes . . . . .

Sediment- Controlled Processes Affecting Water Quality

Microbial Processes . . . . . .

Some Related Concepts from Coastal Studies

Long-term Accumulation .

Short—term Buffering Effect . . . . .
Documentation of the Effects of Wetlands on Water
Quality . . . . . . . . . . . . . . . . . .

Nutrient Studies . .

Sediment Studies . . . . . . . . . . .

Studies of Wastewater Application on Freshwater

Wetlands . . . . . . . . . . . .
Swmmy.

Methods for Estimating the Economic Value of Freshwater
Wetlands in Improving Water Quality . . . . . .

Framework for Valuation .

Economic Tools for Valuing the Wetland Service of Water
Quality Improvement . . . . .

Intermediate Good Method .
Land Price Analysis

Travel Cost Method .
Alternative Cost Method

Applications of Methods to Value Freshwater Wetlands in
Improving Water Quality . . . . . .

Value Studies of Water Quality Effects .
Evaluation of Value Studies
The Potential for Sound Economic Values

vi

Page

85
87
87
87
88

90
91
91
'92
95
95
96
96
96

97
97
98

100
101

102
103

105
106
107
107
108

109
109
110
113

Page

CHAPTER VII: SUMMARY AND CONCLUSIONS .............. 115
Summary ............... . . . . . . . . . . . . 115
Conclusions ................... . . . . . . 116

The Wetland Resource Allocation Decision ......... 117
\ Development Value ................... 117
KPreservation Value . ............... . . . 118
Methodological Issues ................... 118
Implications for Michigan Wetland Resource Policy ..... 119
General Policy Implications .............. 120
Implications for Implementing Michigan Legislation . . . 121
The Role of Science in Natural Resource Policy ...... 122
Recommendations for Future Research . ........... 124

APPENDIX A: COMMON PROPERTY OWNERSHIP AND BEHAVIOR ....... 126

APPENDIX B: PROPERTY VALUES AND BENEFIT MEASUREMENT . . . . . . 132

APPENDIX C: DATA MODIFICATION INFORMATION ........... 158

APPENDIX D: ADDITIONAL INFORMATION ABOUT THE DATA ....... 159

BIBLIOGRAPHY ........ . . . . . . ............ 174

Table

5—2
6—1
6-2

D-l
0-2

[1-3
0-4
0-5
D-6

D-8
0-9

LIST OF TABLES

Theoretical Determinants of Land Value . . ......

Regression Results . . . . . . ......

Processes Affecting Water Quality in Wetlands

Water Quality Values for Michigan Coastal Wetlands . .
Multiplier Adjustment Calculations . . . . ......

Names and Units of Variables . . . . . . . . . . . . .

Clay Township Regression Equations for Vacant and
Occupied Land Parcels . . . . . . . . . .

Clay Township Correlation Matrix . .

Harrison Township East (2a) Correlation Matrix .
Harrison Township East (2b) Correlation Matrix . .
Harrison Township South Correlation Matrix .
Livingston Township Correlation Matrix . .
Description of Joint Test of Significance

Joint Test of Significance--Test 1 .

Joint Test of Significance--Test 2 .

Development Cost Calculations--Clay and Harrison
Township East . . . . . . . . . . . . . .

Development Cost Calculations--Harrison Township
South and Putnam Township . . . . . . .

viii

163
164
165
166
167
168
169
170
171

172

173

 

M . . . . . . . . . . suIfiV hrs-I 'r .':-1:..1 -- "-1’. .1.-.221-1'103r1' £-

 

Figure

3-1
3—2
3-3
3-4

4-1

5-1
6—1
A-l
A-2
A-3
A-4

B-2

0—1

0—2
0-3
0—4

LIST OF FIGURES

The Land-Water Resource Spectrum ........

Optimal Input Mix for Wastewater Assimilation Services . .

Optimal Input Mix for Waterfowl Production Services

Land- Water Resource Spectrum with Wetland Transforming
Technology . . . . . . . . . . . . .

Tradeoff Between Wetlands Preservation and Development
in Balance ..... . . . . . . . . . . . . . . . .

Determination of Developed Wetlands Value
Determination of Natural Wetlands Value
Fishery Productivity Relationships .

Average and Marginal Relationships in the Fishery

Fishery Total Revenue and Cost Relationships ......

Average and Marginal Revenue Curves in the Fishery .
The Hedonic Price Function .

The Marginal Implicit Price Function and Willingness
to Pay Curves . . . . . . . . . . . . . . . ..

Map of Southeastern Michigan Showing Location of
Study Areas . . . . . . . . .

Clay Township Study Area . .
Putnam Township Study Area . .

Harrison Township Study Areas

393
22
28
28

37

53

62
104
127
127
131
131
155

155

159
160
160
161

CHAPTER I
INTRODUCTION

Wetland Resources: An Overview

Perceptions and Values of Wetlands

 

The topic pursued in this research is the use and management of a
natural resource, specifically wetland resources.1 The value of wetland
resources to members of society directly influences the manner in which
these resources are used and managed.; If all individuals were unified
in their preferences and attitudes toward wetlands, their use would not
be an issue. However, perceptions, attitudes, and, thus, values about
wetlands vary widely across the population and also change over time.
There are certain attributes of natural resources like wetlands that
make them particularly susceptible to being valued differently by vari-
ous people. These attributes can be partly revealed by examining the
term "wetland” itself.

f/The term ”wetland“ means "land characterized by the presence of
water at a frequency and duration sufficient to support and that under
normal conditions does support wetland vegetation or aquatic life and
is commonly referred to as a bog, swamp, or marsh” (Goemaere—Anderson
Wetland Protection Act, p. 1, l979)1fi In simple terms, wetlands are
exactly what the word implies, a cSMbination of water and land. Dif—

ferent wetland types may consist of more or less of one or the other

component parts. The two resources are obviously drastically different

 

 

as are the legal and economic forces that control them. For instance,

a physical description of land resources might include such adjectives as
fixed, tangible, and constant with respect to space and time. A des-
cription of the water resource may include words such as mobile, fluid,
changeable, meandering, and the notion of cycling over time. Now com—
bine the notions of the two resources to consider the important attributes
of wetland as a composite resource. Important attributes of wetlands are
not self-evident. Do wetlands have more of land's characteristics or of
water's characteristics? Obviously, they have properties of both as well
as unique characteristics in and of themselves. However, much room is
left for interpretation and selection by individuals as to what are the
important attributes of wetland resources. The perception of different
attributes is complicated even more by the fact that the proportion of
land and water components in a wetland change over time due to natural
forces. Thus, the multi—attribute or schizophrenic nature of wetlands

is the primary reason why people perceive different aspects of wetlands
(land fertility for agricultural use or wildlife habitat) and, in turn,

value them differently.

Wetland Resource Use

 

Differences in values toward wetlands had little relevance when the
resource was only affected by natural factors. Science and technological
innovations of the last century have made it possible for man to use the
wetland resource for alternate purposes. Some of these activities in-
clude drainage for agricultural land, dredging for water transportation,
and filling for residential, industrial, and commercial land.;{For

example, in Michigan, only 3.2 million of an estimated 11.2 million

3

original wetland acres remain (Jaworski and Raphael, pp. 49-50, 19781;}
The lack of understanding of the interdependence of wetlands with other
components of the environment (wildlife, water quality) and the uncer-
tainty of the impact of wetland loss were factors in the wetland conver-

1 The legal system

sion campaign that took place in the United States.
placed little if any constraints on wetland conversion and, in fact,
many government programs directly or indirectly subsidized agricultural
drainage (Goldstein, p. 2, 1971).

The cumulative impacts of wetland destruction eventually became
evident in many areas. This coupled with a greater understanding of
wetland systems by scientists created a new perception of the charac-
teristics of wetlands. Wetlands began to be perceived by some groups
in society as fish and wildlife production areas, flood storage areas,
waste—water‘assimilation areas, and providing many other environmental

services. Thus, the attitude towards wetlands resources changed, and

the value of wetlands in preservation uses changed.

Wetland Protection Legislation

 

The change in values by many groups toward wetlands resulted in
efforts to change the rules regarding wetland resource use. The market
allocation system for wetlands was regarded as unsatisfactory since
wetland public or social benefits were not being considered. Wetland
protection and preservation legislation to regulate wetland use was
enacted at many levels of government. For example, Section 404 of the

1972 Amendments to the Federal Water Pollution Control Act authorized

 

_ lSee Brande (1980, pp. 13—15) for a discussion of wetland conver-
s1on.

the U.S. Army Corps of Engineers to review permits for the discharge of
dredged or fill materials into the nation's waters, including wetlands.
Also, President Carter's Executive Order 11990 (May, 1977) directed
federal agencies to minimize the destruction of wetlands in administer-
ing their programs. Michigan and more than a dozen other states have
enacted legislation specifically regulating wetlands. The Goemaere-
Anderson Wetland Protection Act (Michigan Public Act 203 of 1979) re-
quires that anyone wishing to alter a wetland obtain a permit from the
Michigan Department of Natural Resources. The agency is directed to
balance the benefits and costs of a proposed development as a criterion

for granting a permit.
Economics and Wetland Resources

The role of economics in the implementation of wetland protection
laws deserves discussion. The transformation of any wetland law into
policy requires procedures for determining whether a particular wetland
is worth preserving and for setting priorities among wetlands. As noted
above, some laws offer guidelines for priority setting by requiring that
decisions be based on evidence of benefits and costs of proposed wetland
uses. The argument made here is that economics can provide useful input
into the conceptually correct estimation of those costs and benefits.
Also, many economic concepts are useful for structuring the important

factors in wetland use decisions.
Research Approach

The objective of this study is to construct a framework for the

analysis of wetland use decisions grounded in concepts of economic

06 atoms Inches

    

6 ml: 2mm manor-13m .m-«pu-m «lam 9n.“
1 M95305 sdT .3511“:va 9".Jr-iL-n3'1 '-.;’1'6.-"-' age nei:=::.:n..I

--r-'! WP; 111- :5 ' .-

theory. Specifically, the thesis attempts to develop methodologies for
estimation of the development and preservation values of freshwater wet-
lands. The preservation use considered is the service of water quality
enhancement provided by wetlands. The developed use is residential land
with waterfront location. Once an acceptable methodology is defined,
its application will be attempted. The study region is southeastern
Michigan. Many wetlands in the region are located near water bodies and
thus have potential for providing water quality improvement services.
Proximity of wetlands to water has also made these resources prime areas
for development by individuals desiring waterfront location.

The study is not confined to assessment of physical and biological
data of wetlands and the subsequent calculation of economic values.
Rather, an attempt will be made to place the wetlands resource issue and
the permit decision process in the context of a larger political-economic

framework.

Organization of the Thesis

The thesis is composed of two major subdivisions. The first is the
conceptualization of the wetland resource issue and the formation of
strategies to deal with wetland resource use. This subdivision includes
Chapters II and III. Chapter II presents a general framework for analy-
sis of natural resource issues. Chapter III applies that framework to
the wetland resources of Michigan.

The second subdivision includes ChaptersIV through VI. The focus
of this subdivision is on the allocative decisions about wetlands once
the decision is made to protect them by regulating their use. Chapter

IV specifically states the economic concepts that can assist in wetland

as ”no .musei maker»: rim:

 

at (Io-325'! when etiT ...-aruzre-‘h. as" m..- rruliae" :--1r- 2::

has nit-ea 1935:»: 155a being" 3- -- i"..* '1- ‘ - _ - ' .-.-..-'..-~..=.i

ma Irv-'51.;- 7: .~u:1-'--.--' :1-..

policy decisions. Chapter V presents a methodology for the estimation

of development value of wetlands for residential use. The methodology is
then applied to four separate case study areas in southeastern Michigan.
Chapter VI reviews the state-of-the-art in both biological assessment
and economic assessment capabilities for the estimation of freshwater
wetlands impacts on water quality. The conclusions of the thesis are

found in Chapter VII.

    

.‘u’ um I" «was
W161- ae'Mfl’Pdaqaa Jmeeeau sinuous bus
915 ziaerl: an: in maisufanoa an"? .u'h'z-n. 1535-.- .-.-a chem! amen-m

"- ".' '- fl'

CHAPTER II
A FRAMEWORK FOR THE ANALYSIS OF NATURAL RESOURCE ISSUES

The Political Economy Surrounding Natural Resource Use

What Is the Political Economy?

 

The political economy is the complex system of activities and inter-
actions among significant participants or actors in the political and
economic arena. LAM any given period, the actors in the political econ-
omy will be firms, households, individuals, governments, bureaucracies,
associations and many other units. Each actor has a location within the
political economy which specifies the boundaries which the actor controls
(Shaffer, p. 289, 1969).

The political economy operates through the interaction of partici-
pants. The interactions involve power, status, economic advantage,
resources, technology, standard operating procedures, customs, roles,
and personalities. The units of interaction in the system are trans—
actions which can be social, political, or economic. A transaction will
always involve the transfer of information, rights, privileges, obliga—
tions, benefits, and costs among participants (Shaffer, p. 249, 1969).
Transactions can be classified as bargained transactions, administrative
transactions, or status—grant transactions. Bargained transactions
involve the exchange of rights with the consent of each actor. Certain
rights and rules are acknowledged before exchange occurs. A bargained

transaction usually occurs in a market, and involves both coercion and

 

saU eaves-all In: .1 .21:

7n’hm1r---.' 1?. .mr-g- '-' .'- '-

consent (Schmid, p. 11, 1978). Administrative transactions involve a
one-way exercise or command of authority and utilize coercion. Status—
grant transactions involve a one-way transfer of rights and involve no
coercion. These transactions are governed by roles and customs associ-
ated with social position (Schmid, p. 15, 1978). The extent to which
transactions are bargained, administrative, or status-grant can impact
the performance of the political economy (Schmid and Shaffer, p. 23,
1979).

The transactions that occur as well as the structure and position
of the actors in the political economy are circumscribed by a system of
rules that represent the institutional structure of the political
economy (Shaffer, p. 249, 1969). The system of rules defines opportuni-
ties as well as constraints for participants. The institutional structure

consists of laws, customs, regulations, roles, property rights, and rules.

Basic Issues in the Political Economy

 

The Mix of Political and Market Mechanisms for Preference Articula—

 

tion, Individual participants in the political economy attempt to
achieve their objectives by articulating their preferences in market,
political, and social transactions. The most fundamental issue in the
operation of the political economy is the manner in which the political
and market processes are mixed for eliciting preferences of individual
participants. The combination of political and market mechanisms is
critical because it structures the consequences that have to be taken
into account by participants (Shaffer, pp. 4-6, 19798). The market and
political process should not be visualized as separate entities, but

rather as joint mechanisms for preference articulation. The political

. nasal .nolareoa
-:-.-fiet' .3: .q .M'qua) namm um: am hm

, - w «in Mug-293632 1o .9the-iiai'nimb5 .ifv-Jnl'ep‘tsd 51: -':1"."JE?".-n'

 

.63 .q .191‘sn3 bra 1.861 . um‘n'ne 19-1: :. r": -

system establishes the rules under which the market operates and thus
has a distinct impact upon the outcome (Shaffer, p. 725, 1979A). The
processes of representative government enact the laws, regulations, and
rules that determine the opportunity sets of actors. The opportunity
sets include what participants have to take into account in their deci-
sions. The market can in fact be described as an instrument of the
political system (Shaffer, p. 13, 19798). The implication of this dis-
cussion is that prices determined by market forces express more than the
preferences of individuals acting in their own interest. Instead,
prices reflect politically determined rights and rules concerning what
has to be taken into account in decisions and interactions of those
individuals with rights as they participate in the economy. In short,
the market system generates incentives based on politically determined
decisions concerning rights. According to Shaffer (p. 13, 19798), ”the
political-economic problem is to institute the regulatory system in such
a way that price carries the information and incentive as consistent as
possible with preferences for system performance and the reality of the
environment.”

The preference articulation process has some important characteris—
tics that are worthy of discussion. The market can collect and summarize
an unbelievable amount of idiosyncratic information into the easily and
universally understood fonn of prices. The information is derived from
preferences of participants and environmental conditions (e.g., scarcity)
(Shaffer, p. 13, 19798). The prices generated coordinate production
and consumption activities by providing signals to produce and conserve.

Secondly, the demand revealed in markets is effective demand which

reflects both preferences and ability to pay. In other words, preferences

 

10

receive different weights depending on the amount of income an individual
has. Similarly, preferences that are voiced in the political system are
dependent on the ability of individuals to pay (Shaffer, p. 727, 1979A).
In this case, ability to pay means the ability to generate political
influence. Bartlett (1973) notes that political influence depends on

the ability to provide subsidized selective information to members of the
political process. Therefore, individuals who are able to pay for more
selective subsidized information can make their preferences count more

in the political process. Thus, political power is intertwined with
economic power (Bartlett, p. 156, 1973). Preference articulation is also
affected by the intensity of the interest of the participant group which
depends on how diffuse or specific the benefits of a particular change

in political rules, rights or laws might be. Therefore, we would expect
more intense efforts at political influence by narrow specific interests.
When the benefits are diffuse and smaller to each participant involved,

we can expect less intense efforts (Bartlett, pp. 153—154, 1973).

Interdependence and External Effects. The interaction of partici-

 

pants in the economy produces many complex consequences. Each transaction
that occurs has both internal and external effects. Internal effects are
those consequences which are taken into account by the decision maker.

An external effect is a consequence of an action that is not relevant

to the decision maker given the existing institutional structure. An
effect is external, meaning it is not a part of a participant's decision
making account, if ”it is a profit which cannot be captured or a cost

that need not be borne or a cost or benefit that is not perceived by the

decision maker” (Shaffer, pp. 249—250, 1969).

 

 

11

External effects are ubiquitous. The existence of externalities
reflects the interdependence of individual participants in the economy.
Externalities are often cited as needed to be "internalized". The
approach adopted here does not make such a suggestion. Following Schmid
(1978, p. 10), this approach recognizes the reciprocal nature of pro-
perty rights and externalities. The key public choice question is that
of rights selection that defines what externalities are to be inter-
nalized by what individuals. In other words, what do participants have
to take into account in their decisions?

The primary problem that external effects create for the political
economy arises from the fact that only part of the consequences of the
participant's behavior fall upon him or her. These consequences are
usually the most direct and are thus perceived as the most important
consequences to the participants. The perceived effects are associated
with the behavior and therefore serve as incentives for future decisions
(Shaffer, p. 5, 19798). Participants make decisions based on these
incentives resulting in such externality problems as pollution, public
goods, and the free rider problem. The general problem resulting from
external effects is "that those things that individuals take into account
in making decisions do not result in a performance that is acceptable to
society“ (Shaffer, p. 250, 1969). This phenomenon has been termed a
"social trap" by Platt. The trap occurs when individuals behave in such
a way that increases their individual advantage but cumulatively results
in damage to the society as a whole (Platt, p. 641, 1973).

Two reasons why interdependence and external effects are serious
problems in instituting the political economy are uncertainty and time.

Uncertainty is pervasive concerning the consequences of decisions and

. _ lint Incl blallbe denounce
. our easingouen NDIfl1qGE are: .50. .q are!)

    

JIM! '
at notresnp eaten: fiilgdi V“, ng 265+.r
- - [-71- 1 .__‘ . f _ _ .
Jul-'3'} 5;! '5... 21".: : .- . : I_ _'_l L. x, . f- '1 “...-’15“
I .... ,-- '1‘" ' . . .I.

._‘l

,1-';.:. , ..;

12

actions. The delay in time between actions and consequences makes direct
cause-effect relations difficult to establish (Shaffer, p. 6, 1979B).
Individuals make decisions ignoring these effects resulting in conse-
quences that are not in harmony with overall or longer range goals of

the society.

Natural Resources in the Political Economy

 

Attributes of Natural Resources. The physical attributes of natural
resources generally have many dimensions. The various attributes link
natural resources to other components of the environment. When natural
resources are used or managed, these attributes represent sources of
interdependence among individuals and groups in society. Schmid (1980B,
p. 77) argues that better understanding of the sources of interdependence
and their interaction with rights and rules will lead to better predic-
tion of consequences and possibly new institutional approaches. Schmid
suggests a classification of sources of interdependencies with respect
to resource use. The general classes of sources or situations of inter-
dependence are incompatible use goods, joint impact goods, economies of
scale, transaction costs, and surpluses (Schmid, p. 79, 1980B).

The three categories that apply to the framework developed here for
the wetland case are incompatible use goods, joint impact goods, and
transaction costs. Incompatible use implies that one use of a natural
resource may prevent another use by another individual or group. Scar-
city of the resource results in a situation where only one user can
appropriate benefits from a resource, possibly evoking conflict. Con-

versely, a joint impact good is a good that many individuals can enjoy

 

' m at an m m m
1391302 ad:

13

without diminishing the use by another. The additional cost of another
user is zero. The problem with respect to interdependence is that some-
one must pay for the initial cost of providing the good (Schmid, p. 84,
1980B). Transaction costs are the costs of human interaction and are
unequally distributed in a society. Differences in individual trans-
action costs produce situations where one person's action can affect
anothers. Three types of transaction costs exist. They are contrac-
tual, informational, and enforcement costs (Schmid, p. 88, 1978). These
sources of interdependence are common to natural resource issues and

will be discussed in greater depth in Chapter III.

Scientific Industrialization and Natural Resources. The scientific

 

industrialization process refers to a combination of science-originated
technology with the process of industrialization. Scientific industrial-
ization has released man from the many constraints of the natural
environment, provided opportunities for improvements in the quality of
life, and produced some remarkable achievements. At the same time,
scientific industrialization has substantially changed the relationships
among members of society adding complex new dimensions to the problems
of social organization and operating the political economy. The funda-
mental reason for these problems is that scientific industrialization
reduces man's dependence on the natural environment and increases his
dependence on his fellow man (Shaffer, p. 246, 1969).

The impact of scientific industrialization upon the interdepend—
encies of individuals in a society can be better understood by examining
certain aspects of the scientific industrialization process. Scientific

industrialization often consists of new technology, changes in extent of

 

14

the market, and new operating procedures that create a new pattern of
external effects. A new technology or complex production process operat—
ing in a new field of action or in a traditional field at higher inten-
sity can create new interdependencies and external effects. Scientific
industrialization can result in changes in relative prices, the level

and distribution of income, coordinating mechanisms, and scale of produc-
tion. A new pattern of external effects replaces the accustomed pattern,
thereby providing incentives for individuals to act to achieve protec-
tion or control in the form of new institutions. Adaptable and innova-
tive institutional arrangements are necessary if the performance of the
political economy is to meet socially desirable goals (Shaffer, pp.245-
247, 1969).

The adjustment process to scientific industrialization is never
simple. Individuals attempt to protect themselves from negative effects
and take advantage of potential positive effects. Firms and households
attempt to adjust their span of control to gain potential benefits and
protect themselves from costs imposed by others. Some individuals seek
to voice their preferences for protection by forming associations. The
interaction of the scientific industrialization process with the acti-
vities of participants jockeying for position produces major changes in
the structure and organization of the political economy. The new insti-
tutions that evolve determine the degree to which the external effects
of a new technology have to be taken into account by decision making
parties. Often the institutional adjustments will be followed by another
round of investment in technology, and the process repeats itself

(Shaffer, pp. 247-250, 1969).

    

. . _ whom? an arm-a an em
[not ail: .aefihq evinréw at 25min: at tic-1m as; misertlshnobni
-3ubo1q la afar: one .anzinpnaem Gni'iMib'mo: .-------:-ur - --:: Hui-"Lian! bns

I'E‘u'Fl..r.'.'-.'.':. “1:" .- {I . ... . ' ... u '- ._.. _ .'- .4031.-

15

Scientific industrialization has added a dynamic dimension to the
political economy. The impact is particularly great upon natural re—
sources because of their many attributes which are sources of interde-
pendence. Thus, the use of a natural resource in a new production
process can create new and often uncertain patterns of external effects.
Political and social interaction produce new institutions and property
right definitions. The processes involved are dynamic and complex.

An analytical approach for studying these processes is presented in the
next section.
Analysis of Political Economy of Natural Resources:
The Structure—Conduct-Performance Paradigm
Introduction

The purpose of this section is to present a conceptual framework
for discussion and analysis of activity, structure, organization, and
the resulting performance of the political economy. The industrial or-
ganization or market structure-conduct-performance (S-C-P) paradigm is
the foundation of the conceptual framework used in this chapter. The
paradigm was developed by Edwin S. Mason in the 1930's and later ex-
tended by Joe S. Bain. For further discussion of the market S-C—P
paradigm and its extensions, see Chapter 1 of Scherer (1980). Schmid
and Shaffer modified and extended the S-C-P paradigm to study the poli-
tical economy of a community. In this context, the S-C-P paradigm is
concerned with the relationship between the institutional structures of
a community and the behavior of individual community members in response
to that structure. Further, the framework is concerned with the flow

of consequences or performance of the system as a whole (Schmid and

 

16

Shaffer, p. 6, 1979). The participants in the political economy exist
in a particular environmental situation or structure. The structure
determines their opportunity sets. Each participant reacts to his or
her structural situation with certain conduct. The sum of the responses

results in a change in the environment which may alternatively be called

performance (Shaffer, p. 3, 1979B). Performance provides the benefits
and costs for participants and also serves as an input into an evolving
system (Shaffer, p. 5, 1979B). The essential components of the paradigm

are discussed in greater detail below.

Structure. The structure component of the paradigm consists of all
the characteristics of the environment that constrain the choices of
participants (Schmid and Shaffer, p. 6, 1979). These characteristics
may be physical, technological, ideological, or institutional. Structure
is the system of organization and control of resources in a society and
establishes the matrix of opportunities from which participant must
choose. There are three major elements of structure: jurisdictional
boundaries, property rights, and rules of representation. Jurisdictional
boundaries refer to what is in a community's sphere of control (Schmid
and Shaffer, p. 7, 1979). Property is the relationship among individuals
with regard to the use and control of resources. Property ownership
implies that some participants have a degree of control over a resource
that entitles them to certain benefits. Ownership also involves obliga—
tions which define rights of other individuals in the same resource
(Schmid and Shaffer, p. 12, 1979). The three major ownership types are
private, public, and common property resources. Common property owner-
ship has been frequent for natural resources. Commonly owned resources

can be used by anyone able to physically appropriate them. This

 

17

ownership type leads to inefficient allocation if a resource is scarce

(Schmid and Shaffer, p. 14, 1979). The last important concept related
to property is property rules. Property rules govern what effects have
to be considered in the use of resources.

Rules of representation are a third aspect of structure. Rules of
representation determine the manner in which other rules and regulations
are made and interpreted. These rules determine whose preferences are

counted in decision making (Schmid and Shaffer, p. 15, 1979).

Conduct. Conduct (or behavior) in the S-C-P paradigm refers to the
choices, strategies, and decisions that the participant adopts in response
to the opportunities inherent in a given structure (Schmid and Shaffer,
p. 6, 1979). Economic theory suggests some basic assumptions about human
behavior. Individuals are assumed to act in their self-interest and
attempt to maximize their satisfaction. Consumers do this by maximizing
utility while producers maximize profit. These behavioral assumptions
are often useful in making predictions about behavior.

Since conduct is the linkage between structure and performance, it
is important to have an idea of the objectives, goals, and ideology of
all relevant participant groups to achieve desired results. Important
participant groups in natural resource issues are individual households,
firms, representative government, bureaucratic government, organizations,

associations, interest groups, the news media and others.

Performance. Performance is the flow of consequences of a parti-
cular structure given the behavior of the participants (Schmid and
Shaffer, p. 19, 1979). Performance is a complex concept which refers to

the aggregate result of the choices made by participants. The aggregate

 

18

result can be a change in the environment or the political economic
structure (Shaffer, p. 5, 1979B). Documenting performance can be a
difficult task. Schmid and Shaffer (1979, p; 21) suggest the use of
performance accounts consistisng of performance categories and impact

indicators.

The Dynamic Nature of the S-C—P Paradigm

 

The outcome of one S-C-P sequence contains not only the distribution
of benefits and costs to participants but also serves as an input to the
next sequence. The outcome can change the structure of the subsequent
period. Also, the distribution of benefits and costs can serve as re-
inforcers for certain behavior patterns. In other words, behavior is
shaped by the consequences of actions in each sequence (Shaffer, p. 723,
1979A). Thus, participants acquire new perceptions and preferences in
the learning process which modify their behavior in the following
sequence. The sequential process continues and the system evolves
(Shaffer, p. 3, 19798).

The behavior modification process can shed some insight upon
changing institutions in the political economy. The payoffs to indi-
viduals in terms of outcomes are the contingencies of reinforcement
(Platt, p. 35, 1973). Regulations, laws, property rights, and rules
can change the contingencies of reinforcement by changing what has to
be taken into account in decisions. Laws, regulations, rights, and
rules do this by altering the benefits and costs that are associated
with different behavior patterns (Shaffer, p. 723, 1979A). Therefore,
if a desired outcome requires changed behavior by participants, then
altering the contingencies of reinforcement (laws, regulations, pro-

perty rights, and rules) can help achieve that performance. Stated

 

19

another way, the incentives participants face often require adjustment

to achieve desired outcomes.

The Role of Research

 

The major purpose of research using the S—C-P paradigm is better
prediction of performance or outcomes of alternative structures. Mea—
sures of performance are necessary to facilitate comparison and choice
among alternatives. Choices involve analysis of tradeoffs or the sets
of benefits and costs associated with each alternative (Schmid and
Shaffer, p. 19, 1979). Ultimately, many of these choices will be made
in the political process. A better understanding of the system, includ-
ing more infonnation on the consequences of alternatives, will hopefully

result in improved performance of the system.

    

airman clauses-1 in «mug 10!“ MT
4391 .es-maau'ma sviasngsfls ’re zemoaduo we -_1m:---1‘~r.n 1 nothing“
iiofis uni nnE'“-Jmi_ -iL:?f?‘:i p! v15 ., ' _-' “ ~“wv' lr fined:

71-; _ ' 1n :1fi7-hnw: ‘- ' -v nh --'r n; -: ~' 7 .

CHAPTER III
APPLICATION OF FRAMEWORK T0 WETLAND RESOURCES

Wetlands have become a significant natural resource in the United
States in the past decade. The question that arises is why. The answer
proposed here is that wetland resources, like many other environmental
resources, can be transformed and their nature changed by people through
the manipulation of technology. Therefore, to understand wetland alloca-
tion issues in greater depth, we must understand man's interaction with
the wetland resource. A conceptual model of how, when, where and why
man interacts with the wetland resource is developed in this chapter.

The Structure-Conduct-Performance paradigm presented in Chapter II is
utilized to provide an overall framework for conceptualization of the

problem and as a focus for policy discussion.

The Wetland Resources Model

Structure

The structural aspect of the conceptual framework includes simply
the predetermined characteristics of the setting in which wetland deci-
sions are made. These characteristics may be physical, biological,
environmental, technological and institutional. The similarity among aM
these structural characteristics is that they all act as constraints,
limitations or restrictions upon the choices and opportunities of indi-

viduals within a society. The discussion that follows in this section

20

21

attempts to delineate the important structural characteristics of wetland

resource use, management, and policy.

Physical Structure. To understand the nature of the wetland re—
source issue, discussion and analysis must begin at the most basic level.
An elementary question that arises is ”What are wetlands?” The answer
is that wetlands are a combination of two components of the environment,
land and water. To operationalize this definition, a simple resources
model was constructed as presented in Figure 3-1. In Figure 3—1, land
and water resources are located at the extreme ends of a continuum or
spectrum of various land-water combinations. Wetland resources are
generically defined as the land—water interface that exists between the
land and water resource boundaries at each end of the spectrum. A given
wetland type is identified in Figure 3-1 as a square with a definite
location on the continuum. The wetland type has a particular land-water
mix represented by the position it occupies on the spectrum.

Wetland resources present a unique problem in that their component
parts of land and water are so drastically different from each other.
Land resources are generally of a fixed, defined nature and are more or
less independent from other parts of the environment. Land interacts
with the other components of the environment on a limited scale over
time. Conversely, water resources are generally more dynamic, mobile,
changeable, fluid and less well defined. Water resources interact and
are thus interdependent with the other environmental components. The
distinct differences between the land and water elements of wetland
resources have important implications for wetland use and will be

explored in the next few sections.

 

111.117.
>¢<oz=om

.nwbza’

awk<3

 

 

ESEEEEE “5. Emma >x<azaoa
az<4
azaw_
”353m 52:.
Jupcha .
<_»_z_v
3

mmu_>xum 4<zampxm

Echowam 66L30mwm cmpmzuucwo och
H-m aa=m_a

 

23

4'

Wetland ProductionIhpcesses-Hydr6bgical and Biological Structure.

 

A particular wetland type, depicted as a square on the land-water con—
tinuum (Figure 3-1), can be visualized as a "factory" or production
center which generates a flow of wetland services (Murray, p. 11, 1980).

[The production processes occurring within a wetland factory may be of a
chemical, biological, or hydrological nature:

Several important points are worth noting about the production pro—
cesses in a wetland "factory“. First, there are many inputs into the
production process. Only the inputs of land and water have received
attention thus far. Some other inputs are light, gases and nutrients.
In the naturalenvironment, all these inputs depend upon the forces of
nature. If nature's forces are constant over time, the inputs are con-
stant. If nature's forces vary seasonally or in trends over time,
inputs to the wetland “factory” also vary. In terms of the two major
inputs, land and water, the input-mix and therefore position on the re-
source continuum depends upon whether these inputs vary or are constant
in nature.

The second point concerns the production function which relates
inputs described above to the outputs or flow of services generated.
Like any other ”factory”, the wetland's output depends upon the combin-
ation of inputs used in the production process. Therefore, the flow of
services eminating from a wetland depends upon the land-water input mix
represented by position on the continuum in Figure 3—1.

In discussing the output of the wetland production process, it is
useful to distinguish between external and private outputs. Wetland

external outputs are goods or services that are produced on a wetland

L.»

x ,

  
    
   

  

Inf! I IOUINSRI! «:14» «canon

busirew s ninttw pnf11uoao 29239181n inirsubown edT]

 
 

M11116" {azioulmby‘ we .'-':-":-'I 'r-.! fart-ed.)

ma 3471:1152 - '. .'-;-.

24 " ’1

but refuse to obey the boundaries of the wetland parcelj/ External
outputs flow or spill over into other areas or regions. The outputs are
able to do this because of the many interactions and interdependencies
between wetlands and the rest of the environment. These interactions
provide a medium for transport of the external outputs to other areas
which benefit other individuals and groups.[Examples of wetland outputs
are improved water quality through waste water assimilation, fish and
wildlife via provision of fish and wildlife habitat, reduced flood risk
by floodwater storage, increased groundwater through recharge, and pro-
vision of certain environmental amenities?) Each output has some degree
of the public good characteristics of non:iivalness and non-excludability
(Musgrave and Musgrave, p. 56, 1976). Some of the external outputs can
be termed joint—impact goods because use by one individual does not
diminish the use by another (Schmid, p. 84, 19808).

Private outputs of wetlands are goods or services which have appro-
priable characteristics for which exclusion is possible. Private outputs
are incompatible use goods meaning that use of the output by one indi-
vidual or group prevents use by anyone else. An example is the loca-
tional service of a wetland. One person may be able to occupy a given
wetland location and his presence excludes others from enjoying that
location.

The production of either private or external output exhibits a close
correlation to the land-water input mix. For instance, if the quantity
of land in a given wetland is increased, then the output of private
services increases. This is illustrated in Figure 3-1 by a leftward

movement from the wetland's original position to the square labeled L-1

 

or even further to position L-2, beyond the land boundary. The change
in output of wetland private services corresponding to leftward move-
ment is represented by arrows pointing downward. As these arrows indi-
cate, private services increase in magnitude with leftward movement in
the spectrum. The intuition behind the change in magnitude of these
services lies in the production relationship. As land is used in
greater quantities as an input in the wetland production process, its
characteristics are expressed to a greater degree in the output. Con-
sistent with this explanation, appropriability, which is a characteris-
tic of land, is shown to be an increasing characteristic of the output
of the wetland "factory" as one moves from right to left in the spectrum.

The impact of more or less land or water being added to a wetland
factory upon external outputs is more difficult to predict. At first,
logic may suggest that external services have a simple inverse relation
to private services. If this were true, external services would increase
as water inputs increased in the wetland ”factory”. This relationship
may be true for some wetland services but not for others. For example,
floodwater storage capacity services of wetland may be greater for a
very high water-land ratio. In fact, the service may be most effectively
performed by ponds. In this case, external services increase with
rightward movement in the spectrum and are at a maximum beyond the water
boundary.

Most other wetland external outputs are dependent upon the exist-
ence of some unique aspect of the land-water interface between the
boundaries in Figure 3-1. Production of these outputs cannot occur

beyond either boundary. The unique aspect required for production can

 

26

be certain physical, biological, or chemical conditions that prevail
over some range of land-water mixes. There may be an optimal input
combination that provides the maximum amount of a certain external out-
put type. Assume that this optimal input combination is the center
square in Figure 3—1. Consider a wetland that has a less than optimal
input mix for production of a particular output such as L-1 or W-l.
The output of the wetland will be less than in the optimal input case.
The conclusion is that as one moves further from the optimal input com-
bination, the production of a particular type of output decreases. When
all of one and none of the other input exists (beyond the boundaries),
no production takes place. This relationship is illustrated visually
in Figure 3-1. Arrows pointing upward represent external output of a
certain type. The arrows diminish with left or rightward movement in
the spectrum. A point that should be made is that the production rela-
tionship is not well understood and thus the optimal input combination
for a certain output type is uncertain (Shabman and Batie, p. 13, 1979).
An example may be useful here. An often cited wetland external
output is water quality improvement through waste assimilation. The
land-water interface can create certain conditions that make wetlands
an effective trap or sink for wastes in water. These conditions prevail
over a certain range of land—water input combinations. Within this
range an ideal or optimal input mix exists at which the maximum amount
of waste assimilation services are generated. Our knowledge of the
optimal mix and the production function in general for waste assimila-
tion services is limited. Assume for now that the optimal input mix

is known and can be represented by square A in Figure 3-2. Arrow Z

 

.5552 June: 'rmiiqo en- a: . r‘ : . - '- -- 'L- '1 ir

-: 'r-n‘wn 'lfi.‘ :.: .‘ " ‘ ‘-

27

indicates the maximum output of waste assimilation services. Following
the logic presented earlier, the waste assimilation services decline
with less optimal input combinations (shown by positions to the left or
right of square A). Finally, the external services disappear at, or
near, the water or land boundary.

fflWetlands produce more than one external output, as noted above.
Consider the wetland external output of waterfowl. There are many water-
fowl types utilizing wetland habitat, but for ease of discussion consi-
der only one waterfowl type, say mallard ducksZ} The optimal land-water
mix for mallard duck production is represented by square B with cor-
responding maximum mallard duck production services shown by Arrow Y in
Figure 3-3. This optimal mix for duck production is unlikely to be the
same as the optimal mix for producing waste assimilation services or for
that matter any other external output (including other duck species).
In other words, there may be ranges on a land-water continuum where the
land-water input mix favors the production of one external service more
than another. For example, the optimal input mix for waste assimilation
(square A in Figure 3-2) does not coincide with the optimal mix for
production of mallard ducks (square B in Figure 3-3). Thus, the manipu-
lation of the input mix of a given wetland by increasing land relative
to water (movement from A to B in the figures) will increase the produc-
tion of waterfowl services at the expense of waste assimilation services.
Generally, any changes in the input mix can increase the production of
one external service and decrease another. The important implication of
this discussion is that the land-water mix in a wetland ”factory” not

only affects the relative amounts of private and external outputs but

 

evade belon as ,nvqjuc {smears w- :r' -- ' ~ 1 12'6"?! 351157351! 1

- :-: . 1.111111 9-1-1 awe-i? .. m “1:: ‘.‘ 3"" ' --'

mam-V" "H’. '— "1"

 

Optimal Input Mix for Wastewater Assimilation Services

28

Figure 3-2

 

Wastewater Assimilation Services

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Land Water
Boundary Boundary
Figure 3—3
Optimal Input Mix for Waterfowl Production Services
Waterfowl Production Services
Y
B
Land Water
Boundary Boundary

 

 

 

r-cml

 

29

also influences the mix of external output types. In short, even minor
modification (e.g., diking to lower water levels for duck habitat) may
have to come at the expense of other wetland services (such as waste
assimilation or fish habitat). This tradeoff should be relevant for

management decisions.

Institutional Structure — Property Rights and Rules Concerning

 

Wetland Resources. As noted in Chapter II, property rights and rules
define the means of access to and control of resources in a society.
Property rights and rules can be established by law, custom, or cove-
nant and exist only when recognized by a relevant community. The neces-
sity of acknowledgement illustrates the reciprocal nature of property
rights. Reciprocity means that one person's right is another person's
restriction in the use or enjoyment of a resource (Schmid and Shaffer,
p. 12, 1979).

The three categories of ownership, private, public and common, all
have relevance to wetlands. Wetlands primarily have been owned as pri-
; vate property. Private ownership entails a shared control of resources
1 with government and consists of both rights and obligations. An indi—

‘ vidual has certain rights in his property right bundle but may also have
obligations or restrictions imposed on his use of a resource by govern-
ment. Private owners of wetlands in the past have had relatively few
restrictions on use imposed by government. Rather, the physical nature
of the wetland limited use and fostered impressions that wetlands were
wastelands.

Common ownership or common access implies that rights to use a

resource are held in common by society or by a community. Access to

        

_v'_-*..f‘?"" '

9511111531199 2:; u_:.:_ (:51

393.2 :1n-.‘-2"':':" " '-' I-' "' - '- . .‘ET’

30

commonly owned resources is generally open to all individuals that wish
to use or enjoy the resource. Common ownership can produce efficient
resource allocation and acceptable social results under certain condi-
tions. If the common property resource has sufficient capacity or is in
adequate supply relative to the demands made upon it, then no problem
results from common use. It is unimportant who owns the resource because
enough is available for all. However, if demand for the resource rises,
it is possible for the resource to become overused, degraded and even
destroyed (Seneca and Taussig, p. 93, 1979). This phenomenon was termed
as the “Tragedy of the Commons” by Garret Hardin (1968). The behavioral
reasons why this situation occurs will be discussed in the conduct
section.

5 The importance of common ownership to wetland resources lies behind
thebnature of external outputs produced. Wetlands provide wildlife habi-
tat for waterfowl and fish which have both been treated as common pro-
perty resources. Also, wetlands recharge groundwater and assimilate
wastes thereby improving the quantity and quality of the water resources.
Large bodies of water have been traditionally treated as common property
resources (Seneca and Taussig, p. 73, 1979). The conclusion that can be
drawn is that wetlands provide services that support the existence of
resources that have been historically owned in common. Thus, the re-
sults of common ownership of these resources may be of interest in
analyzing rights to wetland external outputs.y/

Two major points referring to the ownership of wetlands are worth
noting. The first concerns the rights to the inputs, land and water.

Each input has been treated differently by the legal system. Land has

“I
‘1'
.-

107? -' .I1.. '_.

 

31

been primarily owned privately because exCMSion was easy. Water
resources, on the other hand, have largely been treated as common pro-
perty or in cases of degradation and depletion as public property. The
fluid character of water implies that interdependence and interaction
are inevitable and exclusion is difficult. Wetlands, at the land-water
interface, have characteristics of both resources making it extremely
difficult to resolve the issue of what ownership type is most appropri-
ate. As mentioned, wetlands have been principally treated as private
property. Taking this a step further, rights to wetland inputs, land
and water, were also of the private ownership type.

The second important aspect of property rights in wetland resources
involves the ownership of outputs or services from a wetland "factory”.
As pointed out earlier, wetlands' outputs can be classified into either
private or external outputs. Private outputs or services, as their name
implies, are usually owned privately and can be termed incompatible use
goods. Factor ownership controls who is able to enjoy incompatible use
goods like private services (Schmid, p. 79, 1980B). Exclusion is rela-
tively easy. However, the amount of exclusive private services a wetland
generates depends upon the amount of land and water inputs. Thus, nature
also determines the amount of private services wetland owners can enjoy.

Most wetland external outputs do not lend themselves to private
ownership because of the high cost of excluding others from enjoying
them. These outputs or services supported common property resources and
as long as those resources remained in sufficient quantity to meet
demand, little need existed for rights definition in external outputs

of wetlands.

32

The remaining task in defining structure is to identify the signi-
ficant participating groups in wetland resources issues. The simplest
classification of participant groups subdivides all individuals in the
society into wetland private service beneficiames(wetland owners) and
wetland external output beneficiaries. These subgroups can in turn be
broken down into consumers and producers. For now the discussion will
overlook any further subdivision. Other important participant groups

exist but will not be considered in the model yet.

Conduct

The important behavioral assumption stated earlier is that indi-
viduals and firms maximize their satisfaction (utility or profit). For
the purpose of analysis these assumptions are accepted for the partici-
pant groups. Therefore, private output beneficiaries can be expected to
behave in a manner which either increases or protects their access to
the private services of wetlands. External output bemfijciaries can be
expected to behave in a manner that increases or protects their access
to the external services of wetlands.

The conduct of the wetland external output beneficiary is worthy of
discussion. This participant indirectly uses wetland outputs by enjoy—
ing resources that wetlands external services support. The behavior of
concern is the external service beneficiary in common property ownership
situations. Suppose that an individual or firm is currently enjoying
the use of a common property resource. If only a few participants are
using the resource relative to the available supply, their impact on
productivity is likely to be negligible. The high resource productivity

means that each participant is able to earn an economic profit or return

 

33

in this activity. Common ownership presents few problems as long as
productivity and thus economic profit is unaffected by use of the re—
source. Now consider the case where other individuals and firms become
aware of the high return available. Common ownership means free entry
into the activity. New entrants presume that they can earn the average
return in the activity. Consequently, these individuals enter the
activity up to the point where the average cost of using the resource
just equals the average return. At this level of use, the resource's
productivity suffers and all profits or net benefits from use of the
resource are exhausted (Howe, p. 263, 1979). The resource may be de-
pleted or even destroyed under common property ownership arrangements.
The explanation of such a result lies in the behavior of participants.
Each participant acting in his or her own interest only takes into
account the average impact that he or she makes upon productivity. No
single participant has the incentive to take into account the additional
or marginal impact that it makes upon productivity (Nelson, 1979). If
all participants act in the same manner, the collective result will be
that no one is able to enjoy the resource or no effective ownership
exists (Seneca and Taussig, p. 263, 1979). Appendix A presents a

detailed discussion of behavior in common ownership situations.

Performance

The performance or flow of consequences from the structure described
in the preceding pages can be outlined. Natural environmental forces
influence the availability of land and water inputs into a wetland
”factory”. The production process in the wetland translates the inputs

into output of services. Therefore, environmental forces determine the

   

«fin-vs mun! II!” «I: re use

    
 

8W Susana 21051305 well «diving 91!: 03:"
911:! 193W: alsubiu:-bnl and: ..Urwupeano'.‘ .'.L'n'.'l‘33" -:-n".' .-.- "11""-

ormozsri mi:- 1:r:'za=. ‘-'.- '32»

34

quantity and composition of wetland services. Also, since wetland output
types vary with the input mix, the access of participant groups to the
services they enjoy is determined by natural forces. For example, wet
weather conditions usually increase the water component of a wetland.
Private services could diminish while external services may increase.

The main point of this discussion is the dependence of participants on
the natural environment for access to wetland services.

The next important consideration is that many natural environmental
conditions are constant, or proceed at very slow speeds (e.g., geological
time periods), occur seasonally, or in trends. Whether constant, seasonal,
or in trends, the environment's impact on the wetland "factory's" inputs
causes output to vary in similar patterns. Individual participant groups
become accustomed to these patterns and expect them to continue into the
future. However, in a situation where they did not, the only entity to
blame was nature itself, with whom no recourse was available.

As noted, ownership rights to wetland resources have been histor-
ically more akin to private ownership rights (such as those that exist
in land) than common or public ownership. Therefore, an owner has the
ability to exclude others from physically trespassing on his wetland.
However, the physical relationship of the wetland to other parts of the
environment prevents the owner from excluding others from enjoying the
external services produced by the wetland. Exclusion costs are too high.
The implication of the high exclusion cost of external services is that
the owner cannot capture or appropriate external services and sell them
in the market.)nThe result is that the value of the external services

\
will not be reflected in the transfer price of a wetland parcel..’The

 

- -' - ".t-In: "' I'.
' - .11“ .

35

price only reflects the private benefits the owner can appropriate and
trade along with his property. It does not include the value of the
external services that other members of society receive. This is termed
a "market failure" because the assumption of no external effects is
violated by external outputs of wetland resources. Therefore, it would
be expected that the price of a wetland parcel is lower than its actual
value to society.

The next question is the importance of under valuing wetlands in
the private market. The answer seems to be of very little importance,
at least under the present structure. The quantity and mix of both wet-
land inputs and outputs is determined by nature. Wetland owners (pri-
vate output beneficiaries) as well as external output beneficiaries depend
on the environment for access to the services they enjoy. The under-
valuation of wetlands means that private output beneficiaries are able to
obtain access by purchasing and holding land at low prices while they
are dependent upon natural forces for any private services. External
output beneffijaries receive external benefits at zero cost but are also
dependent on nature for access to those services. External outputs can
be of different importance depending on the consequences of common pro—
perty ownership of the resources that external outputs support. As
discussed in the conduct section, common ownership can produce adequate
results if demands made upon the resource are insubstantial relative to
supply. In these cases, wetland external services may not be as crucial
in supporting and maintaining productivity of the resource. On the
other hand, if behavior under common property ownership has resulted in
over-exploitation and degradation, wetland external outputs may be of

critical importance to maintenance of the resource. In conclusion, the

36

importance of external output to the maintenance of resources such as
water and wildlife depends on the degree to which the resource has been
depleted or exploited through common ownership. The success of common
ownership will probably vary regionally due to differences in population
levels and resource demand. As a result, the importance of wetland ex-
ternal outputs are also dependent on regional population levels and

demand for the resource.1

Reconsidering the Model with Wetland Transforming Technology

The factor that adds dynamics to the model is that of technology.
Technological innovation provides the inertia for change through the
scientific industrialization process. Scientific industrialization, as
discussed earlier, is the combination of science-originated technology
with industrialization (Shaffer, p. 245, 1969). The process can result
in complex and roundabout production methods that either operate in a
new field of action or in a traditional field at higher levels of inten-
sity. The following discussion addresses the impacts of the scientific

industrialization process on wetland resources.
Structure

Physical Structure. The physical structure with technology can be
described using Figure 3—4. Science and technological innovation produces

the means for man to transform the nature of wetland resources. The

 

1Another factor that may be important is the successfulness of pub-
lic ownership. Public ownership is often the ownership type that replaces
common ownership when it leads to over-exploitation. The effectiveness
of public ownership depends on the structure of the governmental manage-
ment involved.

 

37

111.111

 

 

 

 

 

     

   

  

33.50.. ESESEOEE do mafia
xuh<a >x<omnmn
3:3 ESOZEE ESQZEE 8?...
”855 52:.
I- .fipct . 1—
_<_»_z_v
7; a-.. m-

 

mug—gum .3336

zoo—0250mm mcmsgowmcmgh UcmFHQZ new: Escpomqm woL30mma Lopez-n:84
e-m eczema

 

38

amount of land and water in a wetland is no longer exogenously determined
by nature. Rather, man has the ability to adjust the position of a wet-
land on the land-water resource spectrum. The adjustability of the land—
water mix is illustrated by arrows that represent possible movement from
the wetlands initial state to new land—water combinations such as L-l or

W-l.

Wetland Production Processes ~ Hydrological and Biological Structure.

 

 

As noted above, the introduction of technology made it possible for the
land and water inputs into the wetland "factory" to be controlled by man.
Since output of the "factory" depends on inputs through the production
function in a wetland, man also has the ability to control the quantity
and composition of wetland outputs. Man is no longer dependent upon
nature for the production of wetland services and is now able to truly
manage the wetland “factory" to produce the output he or she prefers. In
terms of Figure 3—4, the position of the wetland on the spectrum can be
adjusted resulting in different quantities and mixes of wetland service

types.

Institutional Structure. The property rights and rules have essen-

 

tially remained unchanged after the advent of scientific industrialization.
Private ownership has been the usual case with regard to wetlands with
nature limiting the amount of private outputs available. Common property
ownership was predominant for the external outputs of wetlands. External
outputs were also determined by natural forces. If natural forces pro-
duced a steady ample supply of external outputs, common property owner-
ship presented no problems. However, the system of property rights was

unsuited for the changes brought on by the scientific industrialization

 

39

process. In particular, the existing institutional structure was silent
on what rights wetland owners have in transforming wetlands and also on
what rights wetland external output beneficiaries had to the outputs they

enjoyed.

Conduct

The introduction of technology which made possible the transforma-
tion of wetland resources had profound effects on the relationship
between the participant groups in the model. To understand the relation-
ship, it is necessary to examine the conduct or behavior of each parti-
cipant group.

Initially, wetland owners were recipients of private services that
were determined by nature. Technological innovations such as tile-
drainage and bulldozers made it possible for wetland owners to determine
the quantity and type of services produced by their wetland. Only the
private services of a wetland are appropriable and can be traded in a
market with the wetland (Shabman and Batie, p. 1, 1979). Therefore, only
the private services are reflected in the price of a wetland. Rational
individuals that act in their self interest tend to consider only the
value of private services in their decisions. As indicated by Figure
3-4, maximizing the amount of private services may be accomplished by
transforming the resource into land. Further value can be added if the
land created has waterfront location. Thus, in many cases, the wetland
is separated into its land and water components which together give the
greatest value to the individual landowner.

The behavior of private output beneficiaries has important ramifi-

cations for other participant groups. The reason for this is that

 

«arenas-13 arm :11 'I'Jan-s- -_«:

.._.._...'...r,_ .-...
u-l . ' l:

40

wetland owners can transform the wetland in such a way that the relation-
ship between the wetland and the rest of the environment could be changed
or perhaps completely severed. Wetland owners that adjust the position
of the wetland "factory“ beyond the water or land boundaries in Figure
3-4 (L-2 or W-2) cause wetland external outputs to disappear. As a
result, wetland external beneficiaries no longer depend upon nature for
access to the services they enjoy. Instead they are dependent upon
other participants in the system, namely wetland owners. This result is
common to the introduction of technology for ”it is implicit in the
scientific industrialization process that it at once reduces man's de-
pendence on nature and increases his dependence on his fellow man”
(Shaffer, p. 246, 1969).

Rational beneficiaries of wetland external output that recognize
the loss of external services will attempt to achieve protection and
secure access to the external outputs. Members of this participant group
have preferences for external outputs but have no means of articulating
them. Markets are not available for trade of these services because
property rights have not been adequately defined (Shabman and Batie,
p. 1, 1979). Frustrated by the market system, some individuals will
attempt to articulate their preferences through the political system.

Beneficiaries of private output find themselves in a different
situation. Individuals in this participant group are able to effectively
register their preferences in markets for wetland private services (fer-
tile agricultural land or a waterfront lot). A wetland owner interested
in maximizing his own utility or profit responding to the incentives of

the market will often opt for development of his wetland. The decision

41

will not take into account the value of external outputs forgone since

preferences for the output were not expressed.

Performance

The performance of the system has been alluded to in the conduct
section. Technology has permitted wetland owners to remove the limita—
tion on their access to private benefits. They are no longer dependent
upon nature for access to private outputs. Self interested wetland
owners often choose to manage their wetland ”factory" to produce the
maximum amount of private services. The performance thus consists of
increased output of private benefits and a new pattern of external
effects.

At the same time, beneficiaries of wetland external services are
not only dependent upon nature for access to external services. They are
now dependent on wetland owners. Given that rational wetland owners will
convert their wetlands to other uses, the production of external output
will decline. Therefore, performance includes a reduced output of ex-
ternal outputs of wetlands. The wetland owners and private service bene-
ficiaries have gained and external service beneficiaries have lost as a
result of scientific industrialization.

Beneficiaries of external output may or may not realize that they
have lost anything. Whether they do depends on several factors. The
first is uncertainty. Uncertainty about the effects and implications of
decisions is pervasive. Also, lack of knowledge about the complexities
of natural systems makes direct cause and effect relationships difficult
to establish. A third factor is the importance of wetland external ser—

vices to the maintenance of the resource that they support. For example,

 

L1. :5. l " 51-: L19“

42 .

if water resources have become depleted through common ownership, the
impact of the loss of waste assimilation services may be noticeable
since the service was critically important. The opposite situation
occurs when the common property resource has few if any demands placed
upon it and is unspoiled. In such a case, the loss of external services
is less important and may not be recognized by wetland external output

beneficiaries.
The Model as an Evolving System

Performance as an Input to Another Period

 

The performance or flow of consequences of one sequence of
Structure-Conduct-Performance feeds back into the system as input in the
next period. It does this in two primary ways. First, performance can
change the structure or environment in the next period. In this model,
several components of the environment have changed. The amount of wet—
land existing has been changed as well as the quantity and mix of output
types produced. The second major way that the flow of consequencesfrom
one period's Structure-Conduct-Performance sequence influences the next
period is through learning. The consequences entail a certain distribu-
tion of benefits and costs for participants which become associated with
certain behavior patterns. The benefits and costs act as reinforcers for
behavior and modify the behavior in the next sequence. In simple terms,
the participants learn from experience. They may change their prefer-
ences, perceptions, and ideas as a result of this learning. An example
would be useful. Wetland external output beneficiaries might observe
the loss of external services. Some connection between this loss and

the use of wetland transforming technology may be perceived. This

 

43

perception could lead to new behavior in the subsequent period where the
individual attempts to protect himself from the costs of the new tech—
nology. Other participant groups also learn. The groups interact and

the system evolves.

Interaction in the Political Economy

 

The Evolution of New Institutions. The interaction that occurs in

 

the political economy after the introduction of a new technology is com-
plex.JiE8rticipant groups attempt to protect themselves from negative
effects and take advantage of positive effects. Participants attempt to
adjust their boundaries of control (Shaffer, p. 247, 1969). Wetland
private output beneficiaries try to establish their rights to use the new
technology and transform wetlands. External output beneficiaries strive
to have their access to external outputs protected. Each participant
seeks to have his interest protected by the government. Participants
voice their preferences to the government. Out of this interaction
emerges the new institutional structure of the political economy. Pro-
tective institutions are formed which determine the degree to which the
external effects of a new technology are internalized (Shaffer, p. 250,
1969).ZiProtective institutions in wetland resource issues are institu-
tions that determine what effects of the transforming technology are to

be taken into account by wetland ownerslf

Wetland Protection Legislation: A Protective Institution

 

Wetland protection legislation has been abundant in the past decade,
indicating the success of wetland external output beneficiaries in

registering their preferences in the political system. Numerous

44

institutional arrangements are possible for the protection of wetland
external outputs. The arrangements can attempt to change the structure,
conduct, or performance of participants in the system. gSome mechanisms
that attempt to change the structure are property rights reassignment,
rule changes, regulations, taxes and subsidies. (The institutional struc-
ture changes define new contingencies of reinforcement that establish
what participants have to take into account in their decisions. Conduct
can be changed directly by regulations and laws although this requires
high levels of supervision and enforcement;;7Performance can also be
directly changed by setting performance standards which must be attained.
Often the new institutions are complicated and affect structure, conduct
and performance. An example is the Pittman-Robertson Act of 1937

(P.L. 415) which established a tax on firearms to generate funds for the
purchase of wetlands and other natural areas. In this case, structure
is impacted by the tax, providing funds for government to act as a parti-
cipant and purchase wetland areas with the objective of achieving better
performance. The Water Bank program of the U.S. Soil Conservation Ser—V
vice provides another example. The government purchases easements on a
ten year basis from farmers. This could be viewed as a change in
structure. The easement purchase represents a payment for the right not
to develop over a ten year period and the exchange occurs on a voluntary
basis.

The predominant form of institutional change involving wetland
preservation in the last decade has been the permit approach. The permit
process consists of the delegation of authority for rights allocation to
an arm of the government, usually a natural resource management agency.

Individuals desiring to develop a wetland they own are required to obtain

  
   

Ill I'llll uI'Slnpc$I

.aatbledua bna 29x3: .enoirsluesn .aognada sluw

  

dude

flgtldsjze 3581 jn5Mytwoi.i91 in eat ngpyifrr: wen u"'3}h espnad: 91D:

JiuhfiFR .Pfifi'ei“". - -.' J? fiWn“ . -'-r -- T- av 1 r:.5y?_':- - cw
r'H-r. -'- ' ' .-';;. 7
I‘

.l
5
1
i
I

  
    
 

a permit (development right) from this agency. The agency is usually
directed to determine in a balancing process if the proposed wetland
development is in the best interest to society. For example, the Corps
of Engineers' wetland protection program under Section 404 of the 1972
Amendments to the Federal Water Pollution Control Act will only approve
projects where the benefits of the proposed development outweigh the
damages to the wetland resource (Federal Register, p. 37137, 1977).

The permit process can be viewed as case by'case assignments of rights

to develop wetlands, meaning that it is a structural change. Alternately,

the permit process can be considered as direct regulation of conduct.

The Michigan Wetland Protection Act: A Protective Institution

 

The Michigan Wetland Protection Act, also called the Goemaere-
Anderson Wetland Protection Act (1979), established a permit mechanism
for wetlands protection. The act requires that anyone wishing to alter
a wetland must obtain a permit from the Michigan Department of Natural
Resources (DNR). Consistent with the federal program, the DNR is
directed to balance the benefits and costs of a proposed alteration
to determine if granting the permit or development right is in the
”public interest“. Criteria for the balancing process are stated in

the act.

Jurisdictional Boundaries. The chosen jurisdictional boundaries of

 

wetland protection in Michigan are both geographical and in terms of use.
Counties with more than 100,000 people and without a wetlands inventory
are excluded from the act. All wetlands contiguous to the Great Lakes,

Lake St. Clair, or any inland lake or stream fall under jurisdiction of

 

all! Woe daemon-vet: t-ezoumr. en; 9.: n:":.-.nv.ad 5'1" d-xsdw SHUNT,

.lilct: R-I‘.‘C "'u '_.t_> "EH . __-.nr-n'll -' I”. -
-. v: .' '..'|-'
l ..n 7' 3. \

46

the Act. Uses that are exempted from coverage of the Goemaere-Anderson
Wetland Protection Act include agricultural, forestry and mining acti-
vities.

The jurisdictional boundaries chosen can be discussed with refer-
ence to the interaction in the political economy. Bartlett (1973)
suggests that economic and political power are interrelated and that
larger economic units will have more representation in the political
arena as well as the economic sphere. Some empirical evidence of this
unequal distribution of political representation is in the jurisdic-
tional boundaries of the wetland protection act. Participant groups
with sufficient economic resources to generate political influence were
able to have their use of wetland resources exempted from the act.
Other groups with less ability to influence the political process were
not able to do so. The implicit meaning of drawing such boundaries is
that the benefits in all cases of the exempt wetland uses are greater
then the damage costs to the wetland resource. Whether these benefits
always exceed the damage costs is a testable question.

The impact of the jurisdictional boundaries chosen and the ability
of participants to articulate preferences is an important factor. For
private output beneficiaries and wetland owners covered by the act,
their preferences for private output are not easily articulated through
the market. Rather, they must apply for a permit and attempt to justify
their use to the DNR. Those uses and geographical areas not in the
jurisdictional boundaries are still able to effectively express their
preferences for private outputs through the market.

The preferences of wetland external output beneficiaries are ex—

pected to be considered by the DNR in the permit process if the wetland

47

use is covered by the act. If the use is not, beneficiaries of external
output that are affected by that use are unable to express their prefer-

ences .

The Bureaucracy: A New Participant. The Michigan Wetland Protec-

 

tion Act brings a new participant into a critically important role in the
political economy of wetland resources. The chosen mechanism for pro—
tecting wetlands, the permit process, directs the DNR to assign rights

to develop wetlands whenever it is in the ”public interest". The imple-
menting agency, the DNR, is a bureaucratic organization which can have
goals, objectives, and an ideology of its own (Bartlett, p. 21, 1973).
Thus, the behavior of organizations such as bureaucracies is an important
factor affecting the performance of wetland protection policy. Bartlett
(1973, p. 22) suggests that bureaucrats adopt security maximizing beha-
vior.

The behavior of individuals within the bureaucracy will not be
examined here. Rather, it is assumed that bureaucrats are simply inter-
ested in implementing the law and completing the systematic analysis of
the benefits and costs of wetland alterations. The research presented
in subsequent chapters develops a methodology for systematic evaluation
of the benefits and costs and applies the methodology where possible.

The research should provide some insight into the difficulty of the task
that has been handed to the bureaucratic agencies making permit decisions.
The conclusion of Chapter III marks the end of the discussion of
the conceptual framework for analysis of wetland resources in the poli-
tical economy. The point of departure for Chapter IV and the following

chapters is that wetland protection is accepted as an important policy

 

m nts‘fm
. roam '
l {Ilsafri19 a asnl :nsnr ‘
. u'J‘F! “~-
.. . 6 cpn'wd 33A
not:

-awq ' -
«oi methanoom Ivan
... '1 ed'
' I'... = .."-'-"---
..:'_ '6:-:':l':l

:n,l -.
-..v_:.: rs-.'- - "
c . or 15F

 

 

CHAPTER IV
BASIC ECONOMIC CONCEPTS FOR EFFECTIVE WETLAND RESOURCE POLICY

Economics: The Study of Choice

Discussion in this chapter accepts the position that certain groups
that value the natural services of wetlands have been able to articulate
their demands for wetlands protection. Wetland protection for assured
production of natural services is now a policy goal. Before policy im-
plementation, however, some important questions need to be addressed.
The questions include ”How much wetland protection is needed?” and "How

can priorities be established among various wetlands?". These questions

always have to be faced because any policy goal or purpose is subject to

a budget constraint, meaning that choices must be made and priorities
established. The science of economics focuses on the study of the ways
in which society makes choices (Nicholson, p. 3, 1979). Economics can

thus shed useful insight into policy choices and decisions.

The Concept of Value

 

As noted above, economics is the study of human choices about how
to allocate scarce resources among competing alternative uses. Economic
analysis assumes that individuals attempt to make choices in a manner
that maximizes their own satisfaction. Observation of human choice can
lead to determination of the value of different commodities to indiv-

iduals. Individual buyers reveal what they are willing to pay for

49

 

50

commodities by their bids and choices. Similarly, sellers reveal that
they are willing to provide certain quantities of their goods at various
prices. The willingness to sell and buy interact and are reconciled in
the process of market exchange (Haveman, p. 21, 1976). The amount of

. money that is exchanged for the commodity is its price. In a well-
functioning market, the exchange price reflects the value of the parti-
cular commodity to both buyers and sellers. One further point is worth
mentioning. The relative prices that emerge from a competitive well-
functioning market represent the relative marginal valuations that
buyers place on that good. The price of a commodity can be called the
marginal social value or benefit (Boadway, p. 14, 1979). Another way

of saying this is that prices reflect the value of the last unit traded.1

Opportunity Cost

The opportunity cost concept is a simple notion that has far reach-
ing implications for policy decisions. The general notion of opportunity
cost is that any decision to produce a particular product necessitates
sacrificing some other product. The reason why the other good cannot be
produced is that resources are scarce, and if they are used for one
purpose, they are not available for another. The opportunity cost con-
cept is most useful in conceptualizing problems of social choice since
it assists in recognition of available alternatives (Nicholson, pp. 221-
222, 1978). To emphasize the point, opportunity cost refers to the fact
that every option in a particular decision has a price in terms of

opportunities foregone.

 

1The concept of total economic value or all units traded will be
discussed in Chapter V.

 

51

The notion of opportunity cost is an especially useful addition to
the policy area of natural resources where many individuals have concep-
tions that certain environmental purposes have infinite or absolute value.
For example, an individual may consider a wetland to have great value in
its natural state, beyond the need for even any valuation. The notion
of opportunity cost rejects this position on the basis that the cost in
terms of options given up by society may at times be greater than bene-
fits of preservation. Each environmental purpose, including wetlands
preservation, must be compared against what is given up (foregone pro-
ducts) to achieve it (Libby, p. 2, 1981).

The opportunity cost principle can be placed into the framework of
wetland preservation decisions. Suppose a wetland is being considered
for drainage into agricultural use. The opportunity cost of preserving
the wetland in this case is the farmland and crops that are sacrificed.
However, the option of wetlands development also entails an opportunity
cost which is the services the wetland provides in its natural state.

In this manner, the opportunity cost principle can be used to structure
a problem in order to focus on what is at stake for various groups in a
particular situation. The establishment of this structure can help to

facilitate effective choice in policy.

Marginality

Marginality, which is related to the opportunity cost principle, is
another concept that is useful for wetland policy implementation deci-
sions. This concept implies that the marginal or additional effects
(benefits or costs) are the effects of primary importance in determining

the efficiency of the production of a particular good or service. Recall

 

52

that the protection of wetlands for provision of natural services has
been accepted as a policy objective. To accomplish this objective, how
much wetland protection is worthwhile? The marginal principle would sug-
gest to increase wetland protection to the point where the marginal bene-
fits of wetland protection are equivalent to the marginal costs of
protection (which are the opportunity costs for each unit of protection).

Consider, for example, the situation where 100 acres of wetland
' exist in a particular region. Figure 4-1 illustrates the marginal prin-
ciple graphically. The figure shows the marginal social benefit curves
for preservation and development of the wetland acreage. The vertical
axis represents benefits while the horizontal axis reflects the percent—
age of wetlands developed. [:The downward slope of the marginal benefit
curve for wetland development indicates that as more development occurs,
the benefits of developing another unit of wetland decrease. Moving
from right to left, the marginal benefits of preservation curve also
exhibits a downward slope. Each additional unit of preservation is worth
less as more wetlands are preserved. The implicit assumption behind the
curves is that of diminishing marginal utility. The foundations of this
assumption lie in the behavioral observation that as people get more of
a particular good, they place a lower value on obtaining an additional
unit.

The curves in Figure 4-1 can be used to illustrate the marginal
concept. Suppose Qn of the wetland area has been developed. Now con-
sider the case where an individual wishes to develop another unit of
wetland expanding development acreage to Qd. The marginal benefits of

expanding development are greater than marginal benefits of preservation

 

53

Figure 4-1

Tradeoff Between Wetlands Preservation and

Development in Balance

Benefits ($)

.mNmH .mm .a .mCOA ucm xcma ”wugzom

 

 

 

x0 --um>cmmmgm mammco< uCMFpmz .................... xoofi
Rood 68338 88.22 23583 .................... Nb
m
60 o co

 

 

   
   

cowum>cmmmgm to
mpwmmcmm Facwmcmz

($) $114eua8

   

  

pewEQOFm>mo mo
muvwwcmm chwmcmz

 

 

 

......-

54

foregone in such expansion. This will hold true for all units of
development up to point Qe' Here, the marginal benefits of development

just offset the foregone preservation benefits. If development is

{expanded one unit further, then the preservation benefits foregone exceed

(the additional benefits of development. The marginal principle would

advise that development should not expand further than Qe' Also, the
marginal principle suggests that there is a point (Qe) where the trade-

off between the two uses is balanced, or where the marginal benefits

of preservation equal the marginal benefits of development. This par-

ticular mix of the two uses maximizes the net benefits from wetland
use. To see this, consider moving to the left or right from Qe' If
one more acre is preserved, the cost of doing so (marginal benefits

foregone) is greater than the benefits of the additional preservation.

'In either case, the net change in social benefits is negative if move-

‘ment is away from Qe (Park and Long, p. 39, 1979).

The Use of Economic Concepts in Wetland Permit Decisions
Guidelines from Wetlands Protection Legislation

Federal Legislation. The federal government's major regulatory
program affecting wetlands was initiated in the Federal Water Pollution
Control Admendments of 1972, (P.L. 92—500). Section 404 of the Amend—
ments directed the Army Corps of Engineers (hereafter, Corps) to regulate
the discharge of dredged or fill material into the nation's water. The
Corps has had broad regulatory powers since the passage of the Rivers
and Harbors Act of 1899. The act required that anyone wishing to con-

struct or excavate in navigable waters obtain a permit from the Corps.

55

The court case of Natural Resources Defense Council versus Calloway

 

(1975) expanded the definition of navigability to include "all the waters
of the United States whose degradation or destruction could affect inter-
state commerce.“ This broad interpretation essentially brought all wet-
lands under the jurisdiction of the Corps permit program. Regulations
were developed by the Corps to evaluate permits for wetland alterations.
The current policy for permit evaluation is to deny permits for wetland
alternatives that are not in the public interest£::?herefore, the Corps
will not grant a permit for wetland development unless it is found

" . . . that the benefits of the proposed alteration outweigh the damages
to the wetland resource and the proposed alteration is necessary to
realize those benefits” (Department of the Army, p. 37137, 1977). Guide—
lines for this determination include the determination of need for the
project, possibility of alternative locations and methods, the degree

and permanence of positive and negative effects, and the cumulative

effects of the project (Department of the Army, p. 31327, 1975);;7

Legislation in Michigan. The Goemaere-Anderson Wetland Protection

 

Act (Michigan Public Act 203 of 1979) is the state of Michigan's primary

mechanism for the regulation of wetland use. The act specifies that

._....
_,__,_._...._._.—o_..—-—'

unless a person has a permit from the state Department of Natural
Resources (DNR) he or she shall not fill, dredge, drain, or otherwise

maintain or construct a development in a wetland. The act directs the

DNR to only approve permits for activities on wetlands that are in the ;
public interest, and if permitting the activity is necessary to realize (
the benefits from the proposed activity. In order to determine if the

proposed activity is in the public interest, ”the benefit which 1

56

reasonably may be expected to accrue from the proposal shall be balanced
against the reasonably foreseeable detriments of the activity.” Nine
general criteria are cited for consideration in this decision. Four of
these criteria are below. i
(1) The extent and permanence of the beneficial and detri-
mental effects which the proposed activity may have on
the public and private uses to which the area is suited,
including the benefits the wetland provides.
(2) The probable impact of each proposal in relation to the
cumulative effect created by other existing and antici-
pated activities in the watershed.
(3) The amount of remaining wetland in the general area.

(4) Economic value, both public and private, of the proposed
land change to the general area.

Informational Requirements of a Wetland Permit Decision

 

The numerous factors cited in the wetlands protection legislation
above illustrate the tremendous amount of information required for a wet-
land permit decision. To acquire, analyze, and consolidate all this
information into a determination of whether a particular proposed develop—
ment is in the public interest is a difficult task. However, the basic
issue in a wetland permit decision is that of social choice. The disci-
pline of economics can be helpful in structuring the basic elements of

the choice.

The process of comparing the benefits and costs of a particular
wetland development contains elements of marginality and opportunity
cost. Each proposed activity is to be evaluated individually, and its
impact upon both development and preservation benefits assessed. The

marginal, additional, or incremental effect of the proposed activity on

/ natural or development values is of concern in permit decisions. This

 

57

approach rejects the notion of absolute or infinite value. It recognizes
that the cost of total development or total preservation is too high.

Closely related is the concept of opportunity cost. By explicitly
mentioning that benefits and costs must be considered in a balancing
process, the legislative guidelines are acknowledging the opportunity
costs involved. Both approval of a permit (development) and denial
(preservation) have a cost. The cost is what must be given up or sac-
rificed in each case. If a permit is approved, the foregone opportunity
of preservation is the cost of the decision. The use of the opportunity
cost principle can assist in revealing what is actually at stake in a
decision.

The economic concepts discussed above as well as the legislative
guidelines provide an indication of the general kinds of information
needed in wetland permit decisions. First, information is needed on the
value to society of the wetland in its natural state. Secondly, infor-
mation is needed on the value to society of the proposed activity on the
wetland. NatUral wetland value and development value information are
essential inputs into the choices inherent in wetland permit decision
making.

The following two chapters of the thesis focus on the generation

of development and natural (preservation) values of wetlands.

 

CHAPTER V
ESTIMATION OF DEVELOPMENT VALUE

There are three purposes for this chapter. First, a methodology is
presented for estimating the development value of wetlands. Secondly,
the methodology is applied to four areas in southeastern Michigan to
estimate the development value of wetlands for residential use. The
development value results are an estimation of the opportunity costs of
wetland preservation in these areas. Finally, the strengths, weaknesses,

and limitations of the approach and results are evaluated.
Methodological Approach

What is Development Value?

 

Willingness to Pay. The development value of a resource is its
value in a specified use. If the developed resource is sold in a market,
its value is determined by what an individual is willing to pay for it
(Mishan, p. 24, 1976). What a person is willing to pay is revealed by
the bid that he or she makes. The bid is an expression of preferences
for the good or service constrained by the person's income, wealth, and
property rights. Individual willingness to pay can be added, resulting
in total willingness to pay or effective demand (Schmid, pp. 114—115,
1980A).

58

    

Ir'.I-I. .l I
MUM WWI 1O WITMITRS

" . . _ . . I _ .T
' 'I_ 1 out-Jim. .:=:' -- .'~"'- '- ': ._ '1-
- :5 . . ' :-.
II
1.
ll '

59

Development Value Measurement Concepts

 

The Demand Curve and Willingness to Pay. The primary tool used to

 

measure willingness to pay for goods or services is the concept of demand.
A demand curve describes the relationship that exists between amounts of
a good or service a buyer will purchase at various prices holding all
other determinants of demand constant. Other factors affecting demand
include income, tastes, and prices of other goods (Anderson and Settle,
pp. 25-26, 1977). The area under the demand curve for a good or service
up to the quantity demanded can approximate total willingness to pay.
Total willingness to pay is a measure of gross benefits to consumers of
a particular good or service and includes the actual market expenditure
plus any excess amount buyers might be induced to pay. This excess
amount, termed consumers' surplus, can be approximated by the area under
the demand curve above market price. This triangular area represents
the net benefits to consumers of a particular good or service (Dwyer,

et al., pp. 9-10, 1977).

The Impact of Project Size on Benefit Measurement. The size of the

 

proposed development has implications for the measurement of development
value. Development value benefits depend on the amount of land created
by wetland alteration. If the land created is a small (marginal) addi-
tion to the supply of the residential property type, then willingness to
pay for the last unit consumed as revealed by the market price is a valid
benefit estimate (Anderson and Settle, p. 33, 1977). Recall from Chapter
IV that in well-functioning markets, the price that emerges represents

the marginal social value or benefit of that good or service.

 

60

The task of benefit estimation becomes much more complex for large
(non—marginal) additions to the supply of residential land. Wetland
development projects creating a large amount of residential land can
shift supply so that prices and willingness to pay change. In this case,
the resulting lower price does not reflect the benefits of the project
(Anderson and Settle, pp. 33-34, 1977). The welfare effects of the price
change must also be evaluated using the concept of consumers' surplus.
The discussion of benefit estimation in the situation of non-marginal
changes will not be pursued further since this research focuses upon the

benefits of small increases in the supply of residential land.

Consideration of Substitutes. The benefit of development from the

 

standpoint of the private owner is the price received for their developed
wetlands net of costs incurred in development. The social development
value of the wetland is quite different from private value because sub—
stitute locations exist (Shabman and Batie, pp. 5-6, 1981). The social
value of wetlands development is the difference between net private
development benefits of a particular wetland and the net private benefits
of developing the next best alternative that satisfies the same demand.
Only if the development activity would not have occurred in the region
of analysis would private net benefits of wetland development be the
appropriate measure of social development value (Luken, p. 10, 1976).

The definition above arises from the recognition that if a proposed
wetland alteration is displaced by public action (e.g., permit denial),
,the opportunity cost of the displacement may be less than the private

benefits foregone if an alternative exists. In such cases, the true

opportunity cost measure is the ”superior attractiveness" of the

 

61

displaced project over the next best alternative (Steiner, p. 894, 1959).
If the "superior attractiveness“ of the displaced site is due to loca-
tion, this increment can be termed "net locational advantage" (Schmid,

p. 146, 1980A). The magnitude of this increment depends on the employ-
ability of displaced resources in the next best alternative investment.1

The discussion above points to the importance of alternative sub-
stitute sites in estimating development value. Consistent with the
concepts above, the following definition is adopted for use in this study.
The social value of developing a wetland is the difference between the
developed wetlands value to consumers and the value to consumers of the
next best alternative residential site (Shabman and Bertelsen, p. 2,
1978). This definition can be put in the with and without framework of
project analysis. The value of wetlands development is the difference
between the economic surplus earned with the project and that earned
without the project (Shabman and Batie, p. 7, 1981).

The definition of development value can be further clarified by re-
ferring to Figure 5—1. Arrow (a) implies that developed wetlands can
provide a flow of development services that, in turn, produce a change
in economic surpluses through linkage (b). The services may be final
(consumed directly) or intermediate (an input into a final consumer
good). The maximum value of the development exists through linkage (a)

and (b). The key question is whether this total value of development

 

1Employability of displaced resources (or funds) affects social
value of development. If no alternative investment exists, the dis-
placed funds will remain idle and the opportunity cost of displacement
is the full private net development benefits. However, if an alterna-
tive exists the funds could be employed in the alternative. The amount
of funds that are unable to be employed in the alternative remain idle
and are the true opportunity cost of not permitting development (see
Steiner).

 

62

Figure 5—1

Determination of Developed Wetlands Value

 

 

 

 

 

 

 

 

Modified Wetlands Developed
Development Wetlands
(IV) (I)
(a)
V

 

 

 

Development at
Alternative Site

(V)

 

 

 

 

Wetlands Services

--Final
--Intermediate

(II)

 

 

 

Value of Services
from Modified
Development

 

(VI)

Value of Wetlands
Derived Services

 

(III)

 

 

 

Source: Shabman and Batie, p. 6, 1981.

 

Value of Services
from Alternative
Site

 

(VII)

 

 

 

63

can be ascribed to the wetland's development. The answer requires
consideration of alternative ways of providing a similar service flow.
Two alternatives are modified development (Box IV - development with
less damage to the wetland's natural services) and development at an
alternative site (Box V). Calculation of development value is now pos—
sible. If no modified development or alternative site exists, then
development value of the site is the value of derived wetland services
(Box 111). However, if modified or alternative site development is
possible, values will exist in Box VI and VII. The value of wetlands
development is the difference between the values (economic surpluses)
in Box III (with wetlands development) and Box VI or VII (without wet-
land development). The major point of Figure 5—1 is that to the
extent that alternatives to wetlands development exist, the social

development value is reduced (Shabman and Batie, pp. 6-7, 1981).

Focus of Development Value Measurement

 

The specific development value to be estimated in this study is that
of wetlands resources in the use of residential property with waterfront
location amenity. Since residential land is a product sold in the land
market, information is available on individuals' willingness to pay for
the resource in this use, therefore making development value estimation
possible.

In many cases in southeastern Michigan, the ”superior attractiveness”
or "net locational advantage“ of wetlands over other residential sites

is their proximity to water bodies. This proximity has caused wetlands

to be developed (filled) to provide the amenity of waterfront location

 

64

for residential homes. The social value of residential development of
wetlands is therefore the difference between the developed wetland with
the amenity and the next best alternative that can satisfy demand for
the amenity. If no such alternative exists, the development value is
simply the value of the site to consumers net of conversion costs.
Measurement in either case can be accomplished by analyzing the value

of creating the waterfront location amenity. Theamenity is traded along
with the land parcel and, thus, its value should be reflected in the
sale price of the land. However, land prices also reflect various other
attributes. The analytical problem is to isolate the amenity from other
attributes of the land parcel (Shabman and Bertelsen, p. 2, 1978).
Therefore, land price analysis and related analytical techniques are

essential for development value estimation.

Measurement Approach: Land Price Analysis

 

The general idea behind using land or property values as a measure
of development benefits is that development services (proximity to open
space or water) become capitalized into market prices for land. Exami—
nation of land prices can thus be a useful activity in determining devel-
opment benefits. The main features of the two approaches to utilizing
land and property values in benefit estimation are presented in this
section. A more detailed technical discussion of these approaches is

provided in Appendix B.

Land Rent Theory and Changes in Property Values. Environmental

 

characteristics of land such as air quality, noise, or proximity to water

can affect a land parcel's productivity. The differences in productivity

 

65

that result will be reflected in the land rent structure and should
appear in land values provided that no surpluses exist (Freeman, p. 108,
1979). An alternate explanation for observing land values relies on

the concept of a factor controlling access to an environmental amenity.
The factor controlling access to waterfront location is land ownership.
The cost of access to these locations is bid up as individuals attempt
to gain access to areas with the amenity. The bidding produces a rent.
The total benefit of a project creating the amenity is the sum of these
rents so long as no surpluses exist. If surpluses exist, then some of
the increased utility is held by individuals and the summation of rents
understates the value of the project. A typical approach to measure
project benefits (change in rents) is to compare the difference in value
of land with access to the amenity and similar land which has no access
to the amenity (Schmid, pp. 142-143, 1980A).

The change in property value approach to measurement of project
benefitshas two factors that complicate its application. The first,
which has already been referred to, is the condition that all surpluses
be eliminated. The second factor is that of relocation activities on
the part of individuals. For example, some people may move to the
developed site causing rents to fall at other locations. It would be
difficult and costly to trace through all these relocation effects. Two
key assumptions that have arisen to cope with the two factors mentioned
above are competitive markets and perfect mobility of buyers (Schmid,
pp. 144-145, 1980A). If these assumptions are met, surpluses are non-
existent and relocation effects can be ignored. In this case, benefits

can be estimated from the change in the value of the site improved by

 

66

the project (Lind, p. 202, 1973). Freeman (1975, p. 472) points out
that satisfying the two assumptions simultaneously is not likely to
happen in practice. Schmid (1980A, p. 146) notes the restrictiveness
of the assumptions as well.

The following conclusion by Freeman (1979, p. 151) summarizes the
issue. He states that "in general, property value changes can be inter-
preted as benefits only when some mechanism exists to assure that there
are no economic surpluses accruing to households, and when there are no

changes in wages or other factor prices."

Land Rent Differentials and the Hedonic Price Method. Land rent

 

differentials at a given point in time are the input into the hedonic
price method of benefit estimation. The central hypothesis of the hedo-
nic price technique is that commodities are valued by individuals for
their utility-bearing characteristics (Rosen, p. 34, 1974). The hedonic
technique is a method for estimating the implicit (hedonic) prices of
characteristics of goods that are slightly different but still in the
same product class. For example, waterfront residential housing is a
product differentiated by characteristics such as lot size, waterfrontage,
and number of rooms. The first step of the hedonic technique involves
estimation of an implicit price relationship which gives the price of any
product as a function of its characteristics. The estimated coefficients
of the characteristics are their implicit prices. The implicit prices
can be used to estimate the marginal benefit or marginal willingness to
pay for small changes in the amenity level (Freeman, pp. 78-80, 1979).
Also, it is possible for the hedonic price data to be utilized in a

second stage of analysis to estimate the inverse demand or marginal

67

willingness to pay functions for an amenity change (Freeman, p. 110,
1979). For the purposes at hand, only the first stage will be discussed
in this section.

Freeman (1979, p. 152) contrasts the hedonic approach to benefit
estimation with the changes in land value method. The advantage of the
hedonic method noted was that restrictive assumptions concerning mobility
and prices were not required. Conditions needed for use of the technique
include a well-functioning market in equilibrium, the region must be
treated as a single market for housing services, and individuals must
have information on alternatives and be free to choose their residential
location (Freeman, p. 122, 1979).

The application of the hedonic technique to measure the development

value of a small amount of wetlands has been outlined. A hedonic price

function for residential land can be constructed and is illustrated below.

Ph = Ph(W, X1, ..... Xn) (Equation 1)
Where:
Ph = Transfer price of land parcel.
W = Waterfront location amenity.
X1, ..... Xn = Other characteristics of land parcels selected

as explanatory variables.

If Equation 1 can be estimated from observations of various prices and
characteristics, then the price of any site can be calculated with know-
ledge of its characteristics. To estimate the benefits of increasing
the amount of a particular characteristic, it is necessary to calculate
that attribute's implicit price. The implicit price of an attribute is
obtained by taking the partial derivative of the hedonic price function

(Equation 1) with respect to that attribute as shown in Equation 2.

 

aPh/a W = PW(W, X1, ..... Xn) (Equat1on 2)
This marginal willingness to pay equation 2 gives the increase in expen-
diture on housing that is necessary to obtain a parcel with one more unit

of the characteristic of waterfront location amenity holding all else

constant (Freeman, p. 79, 1979).

The estimation of development value of wetlands can now be discussed.

Recall that social development value is the difference between the net
development value of the wetland site and the net development value of
the next best alternative site. The amenity created by wetlands develop-
ment under consideration is that of waterfront location. In order to
focus on the difference in values mentioned in the development value de—
finition above, it is necessary to separate the value of the amenity from
the total value of the land parcel. Thus, the analytical problem is to
isolate the value of waterfront location amenity created from filled
marsh from the other attributes of the land.parcel. The hedonic price
technique can assist in this process by estimating the implicit price
(marginal willingness to pay) of waterfront location amenity. The bene—
fits from small (marginal) increases in the available waterfront location
amenity can be found using the coefficients (implicit prices) of the
amenity in the hedonic price equation (Shabman and Bertelsen, p. 3,
1978). The empirical application of this thesis attempts to estimate
coefficients for Equation 1 from previously developed wetlands and use
those coefficients to predict benefits of a small increase in waterfront

residential sites.

 

2The interpretation of the coefficients in the hedonic price equa-
tion that has evolved in the literature is that they represent the mar-
ginal willingness to pay for characteristics of the site (Small, p. 106,
1974; Freeman, p. 77, 1974A).

 

69

Tools for Operationalizing the Approach

 

Development value estimation via the hedonic price approach
described in the previous section depends on several things. First, a
method is needed to identify the important variables that determine land
prices. Secondly, a process is required to select the most important
determinants of land prices for incorporation into the analysis. Finally,
an analytical method is needed to quantitatively examine the relationship
between the explanatory variables chosen and land prices. This section
explores the tools that can assist in Operationalizing the approach and

thus deals with each of the three factors mentioned.

Theoretical Model of Land Values. A theoretical model of important

 

variables is often useful to the estimation of a relationship. A theo-
retical model was desired that would encompass the important variables
that determine land prices. Previous studies indicate that there are
four major categories of variables that have been found useful in predict-
ing land values. The categories are locational, time-related or histor-
ical, improvements, and amenities (Stull, p. 535, 1975; Shabman and
Bertelsen, pp. 4-5, 1978). The categories were used as guidelines in the
development of a theoretical model of the determinants of land value.
Variables that were considered important to land prices are presented in
Table 5-1. As the table indicates, the number of variables that could
theoretically influence land prices is quite large. Collecting informa-
tion on each of these variables would be difficult and costly. For this
reason, an approach was needed to limit the number of explanatory vari—
ables while still keeping the theoretical model intact. The approach

adopted is described in the next section.

 

70

Table 5-1

Theoretical Determinants of Land Value

 

 

LOCATION FACTORS 1 Proximity to commercial centers.

2. Proximity to parks and open space.
3. Proximity to highway access points.
4 Location on street.

5 Location on water.

HISTORICAL FACTORS 1. Changes in the general price level.
2. The local housing supply and demand situation.
The provision of public services and taxes.

IMPROVEMENT FACTORS 1. Structural characteristics

a. number of rooms

b. age of house

c. garage, swimming pool, etc.
d. seawalls, boathouses, etc.

AMENITY FACTORS 1. Community characteristics

a. income levels
b. racial composition

2. Environmental amenities

a. air quality, water quality
b. waterfront location amenity

 

 

Source: Compiled by the author.

Case Study Design and Selection. The case study approach attempts

 

to control for the effects of extraneous influences on a relationship
that is being studied. One way of controlling for the effect of extra-
neous variables is by matching. If a certain variable Z (e.g., public
services) is related to both observation A and B (e.g., land parcel
prices), then the potential source of bias can be removed by selecting
the observations of A and B so that they are equivalent with regard to Z
(Moser and Kalton, p. 220, 1972). In other words, if careful case study

selection could be used to choose observations that were homogeneous with

 

71

regard to an explanatory variable in Table 5-1, the variable can be
deleted from the analysis. Thus, case study design and selection can
be used to reduce the number of explanatory varibles on which data are

needed, thereby decreasing the difficulty and cost of the analysis.

Statistical Approach. The quantification of the relationship be-

 

tween land prices and its determinants was needed for the purposes of
development value estimation. The analytical tool adopted for this
quantification was the technique of ordinary least squares regression.
Regression analysis may be used to predict a dependent variable (e.g.,
price) on the basis of independent or explanatory variables. The pre-
diction is never perfect so an error term is involved. The principle

of least squares consists of determining the values of the unknown para-
meters in the regression model so as to minimize the sum of the squares
of the differences between observed and predicted responses. The resul-
tant parameter estimates are termed least square estimates (Bhattacharyya
and Johnson, p. 334, 1977).

Two basic problems in application of regression analysis include
determination of the functional form of the relationship between the
dependent and independent variables and the selection of what independ-
ent variables are important enough to be included in the regression modeL
Theory, in some cases, suggests what variables to include (Table 5-1)
and what functional forms might be important. For example, the expected
relationship between land prices and year of sale was hypothesized to be
increasing over time at more than a linear trend. Thus, squared terms
are possibilities. Also, many characteristics were thought to initially

add a great deal to the price of a property but more of eachcharacteristic

 

72

added less up to the point where price declined. This suggested a
quadratic or inverse semi-log relationship. Data analysis was also
performed to resolve the question of functional forms and what variables
were important. The criteria for decisions concerning these two prob-
lems were the t-statistic for statistical significance and R2. The
latter was utilized for the selection of functional form of the relation-
ship between dependent and independent variables. If the inclusion of a
variable increased R2, it was generally included in the equation. Also,
the signs of the coefficients were examined for comparison with ngrjgri
expectations.

The selection of independent variables for inclusion in the regres-
sion model was primarily based upon statistical significance of coeffi-
cient estimates as indicated by the t-statistic. A significance level of
.05 was required for inclusion of variables into the model. The signs of
coefficients were examined for evidence of multicollinearity. The cri-
terion of E2 was also utilized when necessary in the manner mentioned
above. The additional increment a variable added to the R2 of an equa-

tion was an important consideration.
Application of Methodology

The purpose of this section is to apply the methodology developed
above. The procedure utilized and interpretation of results are included.
The section on interpretation of results contains the development benefit

calculations.

73

Procedure
The procedure of the empirical component of the research included
three primary steps: data collection, data modification, and statistical

analysis.

Data Collection. The geographical region chosen for analysis was
southeastern Michigan for several reasons. First, the area's wetlands
are generally located in close proximity to water resources, making
these wetlands attractive for development. The methodological approach
focuses on isolating the value of this waterfront amenity. Second, many
wetlands in the region have been developed for various uses including
residential housing. Thus, the chances of finding an adequate data base
were high. Finally, increasing growth in the region creates the poten-
tial for development pressures on the remaining wetlands, making deci-
sions about preservation or development necessary questions to be faced.
Also, these wetlands are significant for the overall wetland resource
of the state of Michigan.

A data base of developed wetland sites in southeastern Michigan was
obtained from the Michigan Department of Natural Resources. The criteria
developed for case study selection included homogeneity with respect to
external factors (e.g., location) and internal factors (e.g., income,
socio-economic characteristics), that some natural wetlands remain nearby,
and location adjacent to lakes, rivers, or streams. The selection pro-
cess consisted of site visits and consultation with local government
officials. Four case study areas adequately met the criteria and were
selected for study. The four areas are (1) a subdivision in Clay Township,

St. Clair County on the north channel of the St. Clair River, (2) a

74

subdivision in Harrison Township, Macomb County, on Lake St. Clair, (3)
several adjacent subdivisions, also in Harrison Township, and (4) several
adjacent subdivisions in Putnam Township, Livingston County on Portage
Lake (See Figures in Appendix D for maps). Each study area consists of
some properties having waterfrontage on the natural waterbody and many
more having frontage on inner channels.

The data collection process consisted of locating information on pro—
perty transfer prices and individual property characteristics. This
information was collected primarily from county and township tax offi-
cials with help from maps and documents identifying parcels and describ-
ing various property attributes. The available information on each
parcel included year of sale, transfer price, value of improvements
estimated by replacement cost for tax purposes, lot size, occupancy,
location within the subdivision, distance from the natural waterbody,
waterfrontage on natural waterbody, waterfrontage on inner channel, and
subdivision location. This information was recorded for each sale obser-

vation.

Data Modification. The only data modification required involved the
treatment of value of improvements over time. Improvements were valued
by tax assessors using manuals based on costs in the early 1970's. The
procedures used for updating those costs to account for inflation varied
among the tax assessors in the townships observed. A procedure was
adopted to cope with this inconsistency. Cost trend multipliers were
obtained from the Marshall-Swift Appraisal manual. This manual was used
by the assessor for valuing improvements in two (Harrison Township) of

the four case study areas. The multipliers appear in Table C—l in

 

 

 

 

75

Appendix C. Improvements in 1972 costs could be adjusted for time by
multiplying their value by the multiplier. It was desired to bring the
value of improvements to their value in the year of sale. Thus, the col-

lected data were modified by use of the multipliers.

Statistical Analysis and Results. Multiple regression analysis was

 

performed on the collected data.3 Before the actual regression, various
functional forms were tested. The functional form of the relationship
between price and the independent variables was chosen that had the

highest R2.

Where two functional forms worked equally well, both are
presented. The results for the case study areas chosen are presented in
Table 5-2. The coefficients for each equation and the corresponding
t-values (in parentheses) are in this table. The number of observations
(n), the coefficient of multiple determination (R2), and the Durbin-
Watson statistic(D—W) are also shown. The discussion below describes
the equation for each study area.

Clay Township. Equation 1 illustrates the regression equation for
this study area. The data set includes both vacant and non—vacant lots
all having some waterfrontage. Separate regression equations for vacant
and non-vacant lots are in Table D-2 of Appendix 0. Various functional
forms were tested and the equation presented was found most desirable in
terms of the R2 and the significance of the coefficients. The coeffi-
cients have the expected signs. All of the coefficients were significant
at the .010 level. The equation’s Durbin-Watson statistic indicates

that serial correlation is not a problem. The R2 reveals that

 

3The multiple regression was performed using The Statistical Package
for Social Sciences (Version 8).

 

 

u_umruuum cemu~:-:_ngaouu meowuo>gmmao do consazc
.AH-o w_n~hv a xrvcoaa< =_ nuance. mam muwc: ucwEoczmaoE can m:o_uw:_cwu m_auwcm>fi

 

 

 

 

 

 

cacxm. Aceo.gv -- -- -- -- -- -- -- -- A-~.~V -- “mom.n-v floun.av Aooo.qv Aeem.sV Ao~3.qv m:_a>-u
mamoN.~ cNHnm. ch cecg.aaom -- -- -- -- ---- ---- -- ---- huqm.om~H ---- mansmaoooo.-ohmnqaa.a mamac.meq Hwam.aomae- m.nma383~ 68_La Rev
.azk accuse
«acne. -- A-8.H-v AHoo~.Hv ASHH-.~V -- -- -- -- -- -- Amman.v --- Aafim.q V Acma.mv Awmm.n-V A¢c~.nv w=~m>-a
cmo.H «menu. no -- anauaooor n~q~.q nw~.me- -- -- -- --1 a-- -- cmomo.mn -- «enema. mm~.HnAH m.~o«mc- ~.mmawe~o woven Amv
easem-
.m:p comwgcmz
Cu ~ooen. -- -- -- -- Anna.-v -- -- -- -- na-.nv -- -- Amn.o~v Amna.ov Anah.o-V A-e.ev wapm>-u
7/ nno.~ nmnon. Nag -- -- -- -- oN~.aho- -- -- --- -- m~o.~no~ -- --- sm~.~ as.mmm m.~amcMH- m.mgm~mm3 mupLa Aamv
«MNea. -- -- -- -- -- -- A~nm.-v ANNH.WV Aw~m.nv --- ---- -- Amow.8av Am~g.av Aema.o-v Aqmm.ov a=_a>-u
fi~o.~ NNONN. Nag -- -- -- -- -- --. mqq.- omH.Hwn www.ma -- -- -- «mm~.H “no.8om ~.a«m3ng- “.mamnCOm wuLLa RaNV
ammo-
.mzh com.ccmx
L
emcee. -- -- -- -- -- A~w~.q-V --- --- -- --- Aenm.uv Ammm.m-V Amqa.~av Anmm.nv --- Aefim.o-V 632,.U
VmRNa.~ naaea. om -- -- -- -- -- «Sana.e- -- -- -- -- mamm.mm hammooooocr nmncme.fi mmows.nm ---- o.aa~mm- mwhum wwwu
«ax-o m\mm POSZS NNNLmLOS WNHmPOS Hm_ozs .mzem_e24 mzem_o NHdzmz page: hazou owxzo buxom: . ~_=4<> ~34<> ~L66» Lea» uaaoawocfl 8.862;»)
_ _ «ewocoawo .

 

 

Lo mhzm_o~ddmou

mu—amox commmwgmuz

H ~-m aLAML

 

 

,:
I

 

 

77

94.8 percent of the variation in the dependent variable (price) is
explained by the independent variables in Equation 1.

Harrison Township East Study Area. Equations 2a and 2b represent
regression equations calculated for this study area on Lake St. Clair.
The data set includes only non-vacant property and all parcels have
waterfrontage. Two different means were used to measure the amenity of
waterfront location. First, the amenity was measured by waterfrontage
on either the natural waterbody or inner man-made channels. The results
for this procedure are shown in Equation 2a. The second way consisted
of using a general waterfrontage variable and measuring location in a
more continuous manner with a distance to waterbody variable. The re-
sults are shown in Equation 2b.

Again, various functional forms and variables were tested, and the
equations presented provided the best fit in terms of R2 and signifi—
cance of the coefficients. All the coefficients have the expected signs
and are significant at the .05 level. The R2 of Equations 2a and 2b
indicate that the equations explain 77.0 and 76.7 percent of the vari—
ation of the dependent variable (price). The Durbin-Watson statistic
is in the acceptable range in both instances.

Harrison Township South Study Area. The regression results for this
study area are found in Equation 3 in Table 5—2. This data set included
vacant and non-vacant land parcels. Also, the study area consisted of
waterfront lots and adjacent inland lots with no waterfrontage. Water-
front location amenity was measured by a waterfrontage variable and also
a location variable (distance). The location variable was measured with
respect to a base or reference point of the amenity of the adjacent

inland lots.

 

 

 

78

Evaluation of alternative functional forms and independent variables
led to the choice of Equation 3. A predominant problem in this data set
is the existence of multicollinearity among several of the independent
variables (distance, waterfront, and lot size--see correlation matrix
in Tables 0—3 through 0—7 of Appendix D). When variables are correlated,
the coefficient of an independent variable depends on what other inde-
pendent variables are in the model. While the estimated coefficients
may not be individually significant, a relationship may exist between
the set of collinear variables and the dependent variable (Neter and
Wasserman, p. 339, 1974). This problem occurred in working with this
data set. None of the collinear variables is significant at the .05
level. Each of the coefficients does have the expected sign. Conse-
quently, the variables were tested jointly using the Chow test. The
variables were significant jointly and were thus incorporated into the
model (See Tables D-8 through 0-10 of Appendix D).

Equation 3 may have another problem. The Durbin-Watson statistic
for testing of serial correlation lies in the inconclusive region at the
.01 level of significance. The question of whether serial correlation
is causing variances to be biased remains unanswered. The R2 in Equation
3 indicates that 75.7 percent of the variation in price is explained by
the independent variables in the equation. As mentioned earlier, all
coefficients have logical signs.

Putnam Township. The regression equation for the Putnam Township
case study area is Equation 4. This data set consisted of vacant and
non—vacant land parcels all having waterfrontage on Portage Lake, Living-

ston County.

a.
.n
..I

 

79

In general, this equation was the least satisfactory of all
regression equations presented. The variables representing waterfront
location amenity did not show up as very significant. The most signi-
ficant variable was waterfrontage. The log of waterfrontage and lot
size are included even though they are significant at the .3 level.
These variables were included because they improved R2. All the other
variables in the model are highly significant. A distance to the
natural waterbody variable proved to be statistically unimportant.

Many different functional forms were tried, and the equation pre-
sented increased R2 and generated more significant coefficients. Also,
all the signs of coefficients are in the expected direction. The Durbin-
Watson statistic is in the acceptable range. The R2 of the Equation 4

means that 89.1 percent of the variation in price is explained by the

regression equation.‘

Interpretation of Results

 

The results of the regression analysis presented in the previous
section can be utilized to estimate the development benefits of wetland
sites. The purpose of this section is to illustrate how the interpre-
tation of results can yield development value estimates. Development
value estimation for each study area in the next section is broken into
two parts: gross development value estimation and the netting out of

development costs to arrive at net development value.

Gross and Net Development Value Estimation. As pointed out earlier

 

in the thesis, development value estimation depends on the availability
of substitute sites that provide some level of waterfront location

amenity. If the alternative can be identified, then the value of

 

 

80

development is the difference between the developed wetland site to
consumers and the value of that alternative site. If no alternative
site can provide the amenity, then the social development value is the
net private development benefits.

The above concepts can be integrated into the statistical results.
If no alternative exists, then all of the coefficients in the regression
equations, except the value of improvement coefficients, are used to pre-
dict gross development value. If an alternative exists, the coefficients
of the regression equation that are unique to the wetland development
site are used to predict development value. The coefficients of other
variables that can be provided by the next best alternative are not used
for benefit prediction. In this way, the difference between the benefits
of developed wetland and the alternative site can be isolated. This
difference is gross development value. Net development value is found
by subtracting conversion costs. Benefit estimation in each of the case
study areas is discussed below.

Clay Township. No alternative sites providing the amenity of water-
front location were identified for this study area. It seems logical
that a person unable to purchase and develop a wetland might locate else-
where in the area or state. However, it is often difficult or nearly
impossible to identify the alternative. A simplifying assumption was
made that no alternative exists that provides the amenity. This assump-
tion will be evaluated later in the chapter.

When no substitute site exists, all the coefficients, except the
improvements coefficient, can be used to predict gross development bene—

fits. Equation 1' illustrates the important coefficients.

 

81

P = -193279.6 + 35.42695YEAR2 + 83 9278H20WFT (Equation 1')
- 6.17844DISTWB

By plugging in values of the independent variables, the gross develop-
ment value can be calculated. For example, a lot with 75 feet of water-
frontage on a channel 1,000 feet from the natural waterbody will have a
gross development value of $33,570 (1980 dollars). Each coefficient in
the equation represents the expected change in price associated with a
one unit change in an independent variable, holding all other independent
variables constant. Thus, the coefficient of distance to waterbody means
that a one foot increase in distance is associated with a $6 decrease in
price, holding all other characteristics of the lot constant. The same
is true for the other variables. Note the impact of distance on price.
The same parcel described above located 2,000 feet from the waterbody has
a gross development value of $27,390.

Net development value is calculated by subtracting development
costs. These costs were approximated using information from several
dredging contractors in the vicinity. These costs for a lot with 75
feet of waterfrontage and 120 feet of depth are roughly $7,500 (See
Tables D—lland 0—12 of Appendix D). Thus, net development benefits of
a lot 1,000 feet from the waterbody are $33,570 minus $7,500 or $26,070.
The value can be interpreted as the present value of the future benefit

flows from the developed site.

Harrison Township East Study Area. No alternative site for develop-
ment was identified for this study area. Thus, all of the regression
coefficients, except the value of improvement terms, were used for
development value estimation. From Equation 2a, the following coeffi-

cients were taken.

 

82

P = 5003375.1 - 134849.1YEAR + 908.055YEAR2 . ,
(Equat1on 2a )
+ 95.558CLWFT i- 361.180WBWFT — 2.448WBWFT2

By plugging values into Equation 2a', the gross development value of a
lot with 50 feet of waterfrontage on the natural waterbody is $38,530.
The value of the same lot on a man-made channel is $31,360. Net devel-
opment value after subtracting costs of approximately $5,500 (for a lot
with a 180 foot depth) is $33,030 for the waterfront lot and $25,860 for
a channel lot.

Equation 2b can also be used to predict development value. Equation
2b' illustrates the coefficients needed for benefit estimation.

P = 4832515.8 - 130997.3YEAR + 883.44YEAR2
(Equation 2b')
+ 7637.028LNH20 - 679.120LNDISTWB

The gross development value of a lot with 50 feet of waterfrontage on the
natural waterbody is $36,620. This value for lots 500, 1,000, and 2,000
feet from the natural waterbody is $32,400, $31,930, and $31,460. Net
development value is calculated by subtracting development costs of
roughly $5,500 for the lots in this area. The results for this equation

are generally consistent with Equation 2a“.

Harrison Township South Study Area. This study area was unique
among the four areas shown. A subdivision of inland lots with no water-
frontage was located adjacent to several subdivisions having direct
waterfront location. The inland lots were presumed to benefit from
being located somewhat near the waterbody and thus have some level of
waterfront location amenity. Further, it was assumed that the adjacent
land was an alternative substitute site in available supply. Therefore,

the social development value of developing the wetland site is the

\,._

 

'll'Y'lflf‘.

83

difference between the developed wetland sites and the adjacent substitute
sites. Following the procedure of Shabman and Bertelsen (p. 6, 1978),

an index for measurement of this difference in waterfront location amen-
ity was constructed. The two variables used to do this were waterfront-
age and a distance variable. Inland adjacent lots had no waterfrontage
and were assigned as the reference point (or zero) in terms of the dis—
tance variable. The distance of the waterfront lots was measured with
respect to the inland lots. Thus, these two variables were proposed to
measure the development value of the wetland sites when an alternative
had been identified.

The hedonic price technique was used to measure development value.
Since only the two variables representing waterfront location amenity are
of interest, we are interested in their implicit (hedonic) price. The
procedure involves taking the partial derivative of Equation 3 with re-
spect to the waterfront location amenity variables (waterfrontage and
distance). Therefore, the coefficients in Equation 3' can be termed the
implicit price of waterfront location amenities and can be used to

P = 35.08996H20WFT + 2248.287LNDIST (Equation 3')
estimate development value. The gross value of developing a lot with
100 feet of waterfrontage at 1,000, 1,500, and 2,000 feet is $19,040,
$19,950, and $20,600. Net development values are found by subtracting
development costs of roughly $9,500. The net development value of a wet-

land 1,500 feet from the alternative site is $10,450.

Putnam Township. No alternative site was easily identified,

and the assumption made for analysis was that none existed.

 

 

84

Therefore, all of the coefficients from Equation 4 (except value of
improvements) are used in development value estimation. Equation 4'
illustrates the prediction equation. A lot with 60 feet of waterfront-
age and 9,000 square feet of lot size has a gross development value of

2
(Equation 4')

P = 2464735.5 — 67869.578YEAR + 462.07519YEAR
+ 1256.9427LNH20 + 3011.104LNLOT

$25,015. Net development value is approximately $19,015 after subtract-

ing costs of roughly $6,000.

Evaluation of Methodology

The methodology to estimate development value of wetland sites is
not without its weaknesses and limitations. The purpose of this section
is to evaluate the methodology in terms of the accuracy of its approach
and assumptions and to also evaluate the procedure and empirical results.
This discussion should provide some insight into the limitations of the

methodology.

Validity of Assumptions

The validity of assumptions used in the methodology needs to be
adequately dealt with. The major assumptions made in the study are

addressed below.

Availability of Substitute Sites. The assumptions made about alter-

 

native development sites were by far the most critical in the analysis.
In three of the four study areas, the assumption made was that no alter-
native site was available. The accuracy of this assumption needs to be

investigated. An alternative development site was identified for one

 

85

case study area. The assumption made to make analysis possible was that
this location was the next best alternative site. Again, the assumption
may not be an accurate one. The question is if the wetland were not
allowed to be developed, where would the next best alternative location
be? The determination of the next best location is the major weakness of
the approach utilized here. This weakness has implications for develop-
ment value estimation. Development values will be overstated if the no
alternative assumption is made when substitutes,in fact exist. This is
evidenced by the case study area where an alternative was identified.
Development values were considerably lower in these areas. Therefore,
to the extent that substitutes to the development of wetlands exist, the

social development value of wetlands development is reduced.

Assumptions of the Hedonic Price Technique. The necessary condi-

 

tions for use of the hedonic method appear to have been met in each of
the case study areas. The areas were established indicating that a market
equilibrium existed. The land market appeared to be functioning well
based on the numerous real estate companies in the areas. It was

assumed that each study area was part of a larger single market for
waterfront property and that individuals had information on alternatives
and were free to choose their residential site. The methodology only
attempted to measure the marginal addition to the supply of waterfront
property in the market. Non-marginal changes cannot be evaluated with
this methodology unless a second step of the hedonic price technique is

performed that identifies the demand curve (Freeman, p. 124, 1979)..

Assumptions in Development Cost Calculations. The calculation of

 

development cost is presented in Appendix D. The resultant costs

 

86

represent rough approximations. A small survey of dredging contractors
produced some values for separate components of wetland dredging and
filling. An average value was utilized for development cost calcula-
tion based on physical dimensions of the lots in each area. It was
assumed that the total of the cost components was equal to development
costs. In reality, the development costs would vary with respect to
scale of operation, location, and many other factors.

Another related cost that should be included in development costs
are any relevant social costs. For example, if the wetland owner is
flooded and is, in turn, subsidized by the government for his loss, this
is a social cost that should be deducted from gross development benefits
along with development costs to arrive at net development benefits. Local
township officials were questioned as to whether flood risk was signifi-
cant. The answer for each area was that flooding was not an important
problem. Therefore, the social cost of subsidized flood damage recovery
was not included in the analysis. In the case where the buyer was
aware of the flood risk and did not expect any assistance in the event
of flooding, it can be expected that the buyer takes this risk into
account by lowering his bid for the property. There is no need to
adjust for social costs of flood damage assistance in this case.

A similar example is that of pollution through septic systems. If
the property owner is imposing costs on other uses of the lake, this cost
should be deducted from social development value. Septic systems were
utilized only in the Putnam Township study area. Therefore, net devel-

opment value should be adjusted downward to reflect this cost.

 

87

Validity of Empirical Analysis

 

The validity of the procedures used to estimate development value
is crucial to the analysis. The major items to be discussed in this

section include the adequacy of the case study design and data sources.

Case Study Design. A possible flaw in the empirical analysis may be
the use of case study design. Local township officials were asked about
homogeneity of subdivisions and neighborhoods with respect to various
characteristics. The study areas were also investigated. This approach
can be criticized as incomplete for case study selection. An improve—
ment to the approach may be the use of a survey or complete data base
(e.g., census tract) to investigate the characteristics of communities
and facilitate effective case study selection. The homogeneity argument
is especially weak where the study area encompassed several subdivisions.
An alternative to case study design and selection which would also entail
higher data collection costs would be a more complete econometric speci—

fication of the regression model to account for any extraneous influences.

Data Sources and Quality. The use of tax assessor's records pre—

 

sents some problems for application of this methodology. Tax records
were easily accessible at relatively low cost. However, the quality of
the records varied from township to township and over time in the same
township. This problem was less significant for the two study areas in
Harrison Township which had a more sophisticated and consistent data
recording system. The impact of the data source on the results is diffi-
cult to determine. The most probable source of error is in the valuation
of improvements. Assessors often used standard appraisal manuals but

differed in the use of modification factors for treatment of time. To

 

88

cope with these discrepancies between townships, the cost-trend
multipliers in a standard appraisal manual were used (See Appendix C).
The use of tax assessor's records also constrained the variables
that could be included in the regression model. For instance, the data
typically available on the assessork record are a description of the
lot, value of improvements, year of sale of the property, sale price,
value of seawalls, value of boathousesand present owner's name. The
available data restricted the use of many of the variables presented in
the theoretical model. On the other hand, the value of improvements
measure is a composite for many of the variables in the theoretical
model. However, the accuracy of this composite depends on the reliabil-
ity of the assessment system. An improvement for future research might
be the use of a more sophisticated data base which would unfortunately

add to the cost of the study.

Usefulness of Methodology

 

The problems and limitations of the approach have been pointed out
throughout this section. However, the empirical results are generally
statistically satisfactory and shed considerable insight upon the factors
that influence development value. The methodology has weaknesses at
various points yet has distinctive analytical power. The power may be
less in the precision of the estimates than in the reasoning required
to produce sound estimates of economic value. The process is not
straightforward and simple. Rather, the process is complicated and has
several strengths and weaknesses. In part, the usefulness of the
methodology is that it identifies the complexities when they occur,

weaknesses (e.g., the availability of substitute sites) and strengths

 

 

 

 

CHAPTER VI
ESTIMATING NATURAL WETLAND VALUES: THE VALUE OF FRESHWATER
WETLANDS IN IMPROVING WATER QUALITY

Natural wetland values are the values of the external outputs
stemming from a wetland (see Figure 3—1). Examples of these outputs
include the services of fish and wildlife habitat, groundwater recharge,
flood control, wastewater assimilation, and the provision of environ-
mental amenitiesTZZThe values of these outputs are not readily observ-
able because the_Wetland owner cannot exclude others from enjoying the
services and, consequently, cannot exchange them in a market This
phenomenon is often called ”market failure" due to the violation of the
exclusion principle and resultant spillover effects (Haveman, p. 33,
1976).<<Thus, the price of the wetland parcel does not reflect the value
of the external outputs. This creates a dilemma for public decision—
makers since they have no information on the values of external outputs
relative to other commodities. In response to this dilemma, scien-
tists have produced evidence on various wetland functions and services.
Some studies have placed dollar values on these functions.

The purpose of this chapter is to focus on the valuation problem
of one specific wetland output, water quality maintenance services of
natural wetlands in southeast Michigan. The assessment of the value of
this function will proceed in two steps. First, an overview of tMastate
of our understanding of the relationship between wetlands and water

quality is presented. Secondly, a summary of the available economic

90

 

91

valuation methods and their appropriate use is discussed. Finally, the
feasibility of estimating sound economic values for this wetland func-

tion is addressed.
The Relationship Between Wetlands and Water Quality

( The wetland function of improving the quality of water resources

(is often cited in scientific, governmental, and other publications. For

(example, Brande (1980, p. 16) notes that many wetlands "improve water

‘ quality by removing solids and unwanted chemicals.” An Environmental
Protection Agency report (p. 10, 1980) states ”wetlands also act as
natural water purification systems." Legislation such as the Goemaere-
Anderson Wetland Protection Act (1979, p. 2) refers to the wetland func-

. tions of pollution treatment through biological and chemical oxidation
and erosion control via sedimentation. Numerous other references to
this function can be found. In addition, dollar values have been attri-
buted to wetlands for the provision of water quality improvement services
(See Gosselink, et al., 1974; Tilton, et al., 1978; Mitch, 1979). The
purpose of this section is to assess the relationship between wetlands
and water quality based on the available scientific findings for the
northen1lake states. The documentation of this relationship is a nec-
essary prerequisite for the calculation of the economic value of wet—

lands' provision of water quality improvement services.

Wetland Processes Affecting Water Quality

The three categories of wetland types that exist in southeast
Michigan are palustrine, lacustrine, and riverine. Palustrine wetland
systems are relatively isolated from water bodies and consequently the

exchange of water is minimal. Lacustrine wetlands are adjacent to or

 

:~a= welsh to s:iisua a“? nn'Vpiq

”:Jr'dvg serti “P

min-o «mu in ...-ma” amuse armors-am MT

. ‘3 noiir.ul hnsijaw

.4 .lefryweW"xap . li‘an532» n‘ v“'

    

HT

92

in the close proximity to lakes. Riverine wetlands are adjacent to or
near rivers or streams with the principal inflow to the wetland from the
river or stream. The wetlands in southeastern Michigan can be generally
classified into upland or lowland wetlands. Upland wetlands consist of
the eutrophic remains of kettle lakes and are often palustrine. Lowland
wetland systems are primarily associated with streams and rivers (Morris-
son, pp. 19—23, 1979). The discussion in this chapter will be limited

to the performance of water quality improvement services by riverine and
to some extent lacustrine wetland systems. The discussion relies heavily
upon a recent literature survey by Burton (1981).

The general processes by which riverine wetlands affect water qual-
ity is by receiving stream or river overflow in high flow periods. The
wetlands retain water and release the water gradually back into the
stream. The velocity of the water decreases while in the marsh, depoe
siting sediment. Other components of the marsh filter or entrap sedi—
ments also. Deposited with the sediment are phosphorus, toxic organic
substances, and heavy metals. In addition, many other elements are
transformed and released into the atmosphere by wetland processes.
Burton (1981) divides these processes occurring in riverine marshes into
the three categories of plant—mediated processes, sediment controlled
processes, and microbial transformations. The processes are discussed

in detail below.

Plant—Mediated Processes. The water quality of marshes is dramati-

 

cally affected by plant production and eventual senescence and decay.
The major processes are illustrated in Table 6-1. The uptake and storage

of nutrients by wetland plants exhibits much variability (e.g., seasonal)

 

93

Table 6-1

Processes Affecting Water Quality in Wetlands

 

 

I. Uptake and Storage in or on Plant Biomass

A.

Plant uptake and storage of nutrients and non-essential ele—
ments from sediments.

Plant uptake and storage of nutrients and non-essential ele-
ments from water.

Plant uptake and storage of nutrients from air.

Sorption of nutrients and non—essential elements from water
and air.

Form substrate for microbial communities which take up and
store nutrients on plant surfaces.

II. Transformation of Material

C.

Change from inorganic to organic compounds.

Change chemistry of surrounding water by photosynthesis and
respiration resulting in percipitation or solubilization of
constituents.

Uptake and metabolism of organic compounds.

III. ”Filtering” of Material from Water

IV. Release of Elements

 

 

A. Release by leaching to water.

B. Release by decomposition to water, sediments, and air.

C. Release by pumping of gases to and from sediments.
Source: Burton, p. 3, 1981.

 

94

and has definite limits. The general pattern for most nutrients is that
of rapid uptake during early spring followed by slower rates of uptake
which eventually lead to release of nutrients to water by the decay of
plants and leaching of nutrients. The conclusion is that the quality of
nutrients capable of being stored in plants in freshwater riverine
marshes is highly seasonal. The maximum amount of plant storage occurs
during the peak of the growing season (Burton, pp. 2-3, 1981).

The role of plants in removing nutrients such as phosphorus and
nitrogen is a key to the understanding of the relatiOnship of wetlands
to water quality of lakes and streams. These two nutrients are often
the limiting factors in aquatic growth in lakes. The emergent vegeta-
tion in marshes usually does not perform long term storage of nutrients
since these plants decay and release the accumulated nutrients into the
water of a wetland. Woody plants, such as shrubs and trees, can be long
term storage sites for nutrients in wetlands. Rooted plants, in con-
junction with microbial processes, serve as “nutrient pumps” that capture
nutrients from the underlying sediments and recycle them into the over-
lying waters (Burton, p. 4, 1981). Kadlec (p. 441, 1978) notes that
emergent macrophytes in wetlands appear to pump nitrogen and phosphorus
from the sediment to the water on an annual basis. Also, submersed
planktonic and periphytic1 plants remove nutrients from the water in a
wetland, but turnover rates for release tend to be rapid. Still, these
plant forms appear to play a significant role in transforming nutrients
to organic forms that settle out and are buried in sediments (Burton,

p. 5, 1981).

 

1The term periphytic applies to the total assemblage of sessile or
attached organisms on any substrate (Reid and Wood, p. 388, 1976).

 

 

95

Sediment-Controlled Processes Affecting Water Quality. During high

 

flow periods, the sediment being carried by a river or stream are brought
into a wetland. The velocity declines as the water moves in sheet flow
rather than channel flow. The decrease in velocity and the presence of
vegetation promotes the fallout of suspended sediment and particles
(Boto and Patrick, p. 279, 1979). Studies of wetland ecosystems in the
context of wastewater applications indicate that gross removal of sus-
pended solids varies dramatically from a net removal of 97 percent to a
net export of 250 percent. Time appears to be an important factor in
measuring levels of suspended solids. A lag time created by settling
and suspension implies that suspended solids levels cannot be compared
at a point in time (Kadlec and Kadlec, p. 442, 1979). On the basis of
wastewater Studies, many researchers generally believe that natural
marshes remove significant amounts of suspended solids.

Suspended sediments carry large quantities of nutrients such as
phosphorus, organic substances, pathogens, and other compounds. There-
fore, the removal of suSpended solids in marshes is an important process

affecting water quality (Burton, p. 5, 1981).

Microbial Processes. The microbial processes that occur in wetlands
can transform some materials into gaseous form, thereby removing them
from water and impacting water quality. For example, the conversion of
nitrate nitrogen to atmospheric nitrogen (denitrification) has been ob-
served to be a microbial transformation in some marshes. Considerable
evidence exists to suggest that denitrification processes are important

processes under certain conditions in wetlands. In some cases, nitrogen

 

96

fixation may offset the denitrification processes (Kadlec and Kadlec,

p. 440, 1979).

Some Related Concepts from Coastal Studies

 

Batie and Park (1979), in a paper on water quality and coastal
wetlands, presented a categorization of functions that wetlands may per—
form which improve water quality. The general logic of the framework
seems applicable to the discussion of freshwater wetlands in Michigan.
The categorization included a long-term accumulation function and a

short—term buffering function which are briefly discussed below.

Long-term Accumulation. As previously mentioned, the velocity

 

reduction and wetland vegetation causes sediment to fall out of suspen-
sion in runoff entering a marsh. If the sediments are permanently
remoVed by burial in the wetland, there may be reductions in turbidity
and siltation in downstream areas. Boto and Patrick (1979, p. 484)

note some related impacts such as increasing dissolved oxygen levels by
reducing the impact of high carbon sediment upon biological oxygen
demand and improving reaeration of streams. Also, if nutrients, pesti-
cides, and other toxic compounds are adsorped to sediments, they may
also be permanently buried with the sediment. In wetlands ”where little
reworking of sediments occurs, deposition can result in virtually per-

manent removal of most pollutants“ (Boto and Patrick, p. 484, 1979).

Short—term Buffering Effect. The second category of wetland func-

 

tions discussed by Batie and Park is that of short—term buffering effect.

This function suggests that coastal wetlands may smooth out or buffer

the time pattern of pollution loading and consequently reduce the

 

97

 

variance of pollution loading levels. The hypothesis of the short—term
effect arises from recognition that sediment accretion in coastal wet-
lands occurs at very high rates for a short duration during storm

events. Therefore, the attention that has been placed on average annual
rates ignores water quality problems due to excessive short-term pollu—
tant loads (Batie and Park, p. 8, 1979). The short-term buffering hypo-
thesis for coastal wetlands may have implications for freshwater riverine

wetlands that also receive their sediment loads during flood events.

Documentation of the Effects of Wetlands on Water Quality

 

The objective of this section is to present the documented evidence
of natural freshwater wetlands impacts on water quality. Midwestern
studies are discussed first because they more closely approximate the
Michigan situation. Studies from other areas are discussed later in the

section if they are believed to be of relevance to freshwater wetlands.

Nutrient Studies. The bulk of knowledge of the effect of riverine
marshes on water quality comes from two studies in Wisconsin (Horicon
Marsh and Theresa Marsh in southeastern Wisconsin). The marshes had
similar hydrologic and nutrient loading regimes. The general pattern
of nutrient uptake and storage was summer through fall uptake and spring
release during high flows. Specifically, the marshes appeared to be a
sink for nitrogen, and Theresa Marsh removed 50 percent of the incoming
phosphorus annually. \The results on nutrient removal were obtained
using input and output concentrations and neither study incorporated
hydrologic inputs and outputs into the study. Mass-balance input-output

are much preferred to determine whether a marsh is a source of sink for

 

 

98

any nutrient or substance. Unfortunately, mass-balance results needed
to quantify water quality effects are largely unavailable (Burton,
pp. 2—6, 1981).

The sparse knowledge of nutrient dynamics and water quality impacts
of freshwater marshes is illustrated in a review by Whigham and Bailey
(1978, pp. 471-475). These authors state that few data exist from mass-
balance studies and that such studies are needed to determine if wetlands
annually accumulate or lose nutrients. The only tentative conclusion
the authors were willing to make was ”that wetlands which have organic
substrates have vegetation that annually accrue smaller amounts of nitro-
gen and phosphorus than does vegetation in wetlands with inorganic sub-
strates."

A review paper by Van der Valk, et al., (1979) summarized wetland
nutrient studies from around the world. All 17 studies indicated that
wetlands improve water quality to some extent by removing phosphorus
and nitrogen at least on a seasonal basis. The efficiency of nutrient
removal varied considerably from site to site due to unique characteris-
tics of the wetland studies (hydrologic regime, vegetation, etc.). The
consensus of this and other studies indicates that ”marshes appear to
be at least seasonal sinks for both nitrogen and phosphorus and are prob-
able sinks on an annual basis for both for many wetlands” (Burton, p. 7,

1981).

Sediment Studies. The only studies of sediment accumulation by
natural freshwater wetlands in the midwest are from Illinois. Burton
(1981) cites a study in southeastern Illinois where wooded wetlands were

found to be a net annual importer of phosphorus sediment. However, in

99

flood periods little phorphorus was removed. Still, the wetlands improved
water quality in low flow periods.

A second study cited by Burton is that of a floodplain forest
(Momence Wetlands) on the Kankakee River in northeastern Illinois. The
results of this study revealed that sediment containing phosphorus was
deposited in the wetland. Burton concludes that "flood plain forests
have considerable value in trapping sediments and nutrients."

Boto and Patrick (1979) presented a paper on the role of wetlands
in removing sediments from water. They note that most research regarding
sedimentation has been confined to estuarine systems, and, thus, few data
exist on either hydrologic flows through freshwater wetlands or sediment
accretion rates. Further, the authors state that "any discussion of the
relative efficiencies of pure grassland systems and forested freshwater
wetlands in promoting deposition by the filtering action of plants would
be speculative."

Sediment removal has been studied to a much greater degree in
coastal wetland systems. The results indicate the sediment accretion
rates vary widely depending on location, the input of sediment, eleva—
tion tidal regime, storm surges and sometimes man-made impacts (Boto
and Patrick, p. 484, 1979). The relationship of sediment removal to
burial of nutrients, pesticides, heavy metals and other toxins has been
mentioned in the literature. There are data available on the affinity of
nutrients and toxics for sediments, but Boto and Patrick (1979, p. 484)
acknowledge the fact that much discussion on the role of sediment—
nutrient removal is "speculative" at this time. Still, Boto and Patrick
present some evidence by DeLaune (1978) of input rates of nitrogen and

phosphorus due to sediment accretion.

 

100

The hypothesis of short-term buffering effects of coastal wetlands
may have some merit according to Batie and Park (1979). The authors
cite evidence by Correll (1979) that the variance associated with the
phosphorus load into a tidal wetlands system was greater than the vari-
ance of the mean outflow of the wetland. The authors also note in Boto
and Patrick's (1979) review references to the ”buffering effect" of

vegetation which could increase sediment accretion.

Studies of Wastewater Application on Freshwater Wetlands. The study

 

most relevant to riverine wetlands in Michigan is that by Fetter, et al.
(1978) on the use of a natural marsh (Brillion Marsh in Wisconsin) for
the purpose of wastewater polishing. Based on an input-output concentra-
tion measurement, they concluded that significant removal of biochemical
oxygen demand, chemical oxygen demand, turbidity, phosphorus, nitrate,
and coliform bacteria occurred in water passing through the marsh.

Kadlec and Kadlec (1979, pp. 449—452) state that in some circum-
stances wetlands can renovate secondarily treated wastewater. They cite
the success of such a project on a partially forested wetland in
Bellaire, Michigan. The authors note that much is to be learned about
uptake and cycling process under changed water quality conditions. Water
quality in wetlands varies over time and flow distance, making measure-
ments difficult and requiring longer term studies to develop sound con-
clusions. The authors emphasize the need for more mass-balance studies
and detailed hydrologic information in order to accurately interpret
wetland function.

Wingham and Bayley (1979, p. 476), in their discussion of nutrient

cycling in freshwater wetlands, further cite the need for mass-balance

 

101

studies. They note that some woody plant-dominated inorganic substrate
wetlands may assimilate and store nutrients. Whigham and Bayley acknow—

ledge "there are, however, not enough data to make definite conclusions.

Summary

A short summary of discussion in the section may be useful. Many
researchers believe that freshwater riverine wetlands substantially im-
pact water quality and have an idea of the processes at work. However,
as Burton (1981, p. 10) notes, "data are too sparse to draw firm conclu—
sions as to whether they are sources or sinks for most nUtrients. The
few studies that have been done do not include the hydrologic data nec-
essary to make such calculations.” Preliminary data results from an
ongoing study of Burton (at Pentwater Marsh, Michigan) indicate that
riverine marshes do remove suspended material and thus impact water
quality. An example of the measurement difficulties created by the com—
plex hydrologic regimes of wetland systems is the problem of lake
seichesz in Burton's study. The seiche activity from Lake Michigan has
created problems for measurement of output from the marsh.

Freshwater wetland processes of sediment and nutrient uptake have
been relatively lightly studied in the quantitative sense in other parts
of the United States. However, wastewater application studies do indi-
cate that wetlands remove pollutants from wastewater, although the pro-

cesses are not well understood.

 

2A seiche is a general term for one form of periodiccurrent system.
It can be described as a standing wave in which some stratum of water in
a basin (lake or estuary) oscillates about one or several nodes. The
oscillating motion can be observed in a bowl with a liquid as it is
passed (Reid and Wood, p. 179, 1976).

 

 

102

Studies of coastal wetland nutrient removal have been more numerous.
Sediment accretion, which is believed to be closely related to nutrient
removal, has been shown to vary widely depending on factors in the study
area. Coastal wetlands also appear to uptake and store nutrients and
sediments.

The importance of risk in extrapolating from studies in other geo-
graphic areas, wetland types, and hydrologic regimes is noted by Burton
(1981, p. 6). He states that extrapolation from a limited geographic
region (e.g., the Wisconsin studies) and range of hydrologic and nutrient
regimes may lead to ”erroneous conclusions“ about how marshes function
in one particular area. Thus, it appears that the complexity, variabil-
ity, and lack of quantitative data are severe constraints on our ability
to understand the relationships between wetlands and water quality in
Michigan.

Methods for Estimating the Economic Value of Freshwater

Wetlands in Improving Water Quality

The recognition that the public and social values of wetlands are
not adequately expressed in private markets has prompted many individuals
to call upon economists to document these values (Brande, p. 16, 1980).
The purpose of this section on economic valuation methods is to review
the valuation concepts and methods that are capable of estimating the
value of wetlands in improving water quality. The basic premise on which
this discussion relies is that wetlands have value because they provide

goods or services thatare valuable to people.

 

 

103

Framework for Valuation

An economic framework suggested by Shabman and Batie (1981) is I
adopted for discussion of valuation.“ Natural wetlands can yield a flow ikf
of services which are valuable to man.f These service flows can be
final, intermediate, or life support as shown in Figure 6—1. Final
services are used directly by a user such as improved water quality for
swimming or home use. Intermediate services are the services which are
inputs into the production of a final product (e.g., improved water
quality as an input into manufacturing of a product). A third category
is life support services which could include nutrient cycling or gas
exchange with the atmosphere. Arrow (a) in Figure 6-1 implies a rela-
tionship exists between a particular wetland's size and characteristics
(e.g., land-water mix) and the levels of wetland service flows (Box I
and II, respectively). Arrow (c) in the figure implies that a quanti-
tative relationship exists between measurable levels of wetlands sub-
stitutes and levels of services. These general types of substitutes for
wetland services are institutional rule change, technological substi-
tutes (waste treatment plants), management, and construction of new wet-
land areas. The linkages (a) and (c) in Figure 6—1 suggest that the
amount of natural wetland services to man at any given time and location
will depend on the combination of wetland substitutes and wetlands area
available. Arrow (b) indicates that the wetlands services affect
people's welfare and thus have value to them. The value should be ob-
tainable either in market prices or by research to determine the ”shadow
values” of services with nonmarketable characteristics. Also, wetlands

substitutes have a cost which is represented by linkage (d).

 

 

 

Determination of Natural Wetlands Value

104

Figure 6-1

 

 

 

 

 

 

 

Wetlands Substitutes

 

 

 

 

 

 

 

)'

 

 

 

 

Wetlands . .
. -— Inst1tut1ons (I)
SUbE§;EUte A; (d) -— Technology (T)
T -- Management (M)
-- Construction (C)
(V)
(IV)
(C)
Economic Wetlands Services
Value of
Service -- Final
Vector i (b) -- Intermediate 4
‘ -- Life Support 7
(III)
(II)
Source: Shabman and Batie, p. 4, 1981.

 

Wetlands Area

-— Size

-- Other Physical
Characteristics

-- Quality.

(I)

 

 

 

105

The "with and without” project analysis principle can be utilized 1
for valuation of wetland services in this framework. If no substitutes
exist which can provide the wetland service, the foregone economic value
of services (Box III) associated with a small change in area of wetlands
can be compared to the value without the change to determine the value
of the wetland. The amount of value can be entirely attributed to the
wetlands area change through arrows (b) and (a).

The case where wetland substitutes exist is more complicated. Here,
the alternate measure of value is the cost of employing the substitute
to replace lost wetlands which can be traced along arrows (c) and (d).
The wetlands value will be the smaller of the cost of wetlands substi-
tutes which supply the same wetlands services and the direct measure of
services that are attributable to the wetland. Shabman and Batie (1981,
p. 5) point out the need for valid biological data and analysis to
estimate sound economic values. One reason for valid data is to define
linkage (a) (economic value of natural services). The services produced
by a given wetland need to be defined for this value to be calculated.

A second reason why biological data on services is needed is to be able
to identify linkage (c) (the substitutes that supply the services) and

linkage (d) (the cost of those substitutes).

Economic Tools for Valuing the Wetland Service of Water Quality
Improvement

 

The presumption of the discussion in this section is that biologi-
cal data and analysis are available such that the services stemming from
a wetland site are known with certainty. In other words, assume the

wetland production function is known. If water quality improvement

 

106

services are among the outputs of a wetland site, its value can be
estimated by use of several tools. The tools used to “shadow value

wetland services are discussed below.

Intermediate Good Method. The intermediate good method, also called
the derived value method, is based upon the contribution of the wetland
service as an input to a final product which sold in a market. The
method is based upon the public intermediate good case where the value
of the good can be indirectly observed in the market. If the profit of
a firm using the input can be observed, then so can the return to the
public intermediate good. Extensive information on the firm's operation
is required, often making application troublesome (Boadway, p. 88, 1979).

An example of the intermediate good method for water quality improve-
ment services would be the evaluation of the input of wetland water
improvement services into the production of a manufacturing firm. The
share of the value attributable to any factor of production (wetland
service, land, capital, and labor) can be found by subtracting the
opportunity cost of all other factors from the market value of output.
The remaining residual represents the return that can be imputed to the
remaining factor (e.g., wetland services) used in production. Utilizing
the with and without principle, the value of the services to be lost in
a proposed wetland development can be estimated. If the change in water
quality associated with the development is marginal and thus does not
affect prices, then the value of the wetland's provision of this service
can be evaluated. This value is equal to the residual return to the
water used in production with the wetlands minus the residual return to

the water without the wetlands in question. A re—examination of Figure

 

 

107

6—1 shows that another step is necessary. The existence and costs of
substitutes to the wetlands services must be considered. If no substi-
tutes exist, no further analysis is needed. If substitutes do exist,
then the lesser of the substitutes cost and the direct measure of wet-
land service benefits is the value of the wetland service. This example
parallels the discussion of coastal wetland fish habitat as an input

into fish production by Shabman and Batie (1981, pp. 7-8).

Land Price Analysis. The price of any particular land parcel is
determined by characteristics that affect its productivity and/or desir-
ability. A site may benefit from a wetland service. For example, a .
residential site may benefit from open space amenities of wetlands. It
is possible that a waterfront residential site may benefit from improved
water quality resulting from wetlands effects on water quality. Parcels
receiving such benefits should be bid up and command higher prices than
parcels not receiving such benefits. An analysis of land sale prices
can reveal the value of the wetland service. Freeman (1979) discusses
this approach in detail. The approach lends itself more readily to
valuation of development services. Since the use of land prices is
discussed in the development value estimation in Chapter V, the method
will not be elaborated on here. The with and without principle and con-
sideration of substitutes are crucial factors in the application of this

method to natural wetland values.

Travel Cost Method. The travel cost method is a widely accepted
approach to estimating the benefits of recreation sites which can be

relatively easily applied (Dwyer, et al., p. 79, 1977). The method

 

108

observes the travel behavior of individuals and attempts to construct

a demand curve for recreation sites. The value of wetlands in providing
recreational services could be evaluated using this method. The water
quality services of wetlands can impact the quality of recreation at a
particular recreational location. Freeman (1979, p. 229) assesses the
possibilities and potential problems of applying the travel cost method
to estimate the benefits of water quality improvements at particular

recreation areas or water bodies.

Alternative Cost Method. The essence of the alternative cost tech-

 

nique is that the estimated value of wetlands in producing a particular
service is approximated by the cost of the next best alternative way of
providing the same service. The alternative cost technique is valid only
under a certain set of conditions. First, the alternative must be cap-
able of satisfying the same demand by supplying the same services that
the wetlands provide. Second, there must be evidence of effective demand
for the alternative project so that if the wetland services were not
available. society would demand and pay the price of the alternative.
The second condition is often of critical importance. If the next best
alternative would not be built then benefit estimation requires know—
ledge of the demand curve for wetland services (Dwyer, et al, pp. 185-
186, 1977). Third, the next best alternative chosen for comparison
should be the least cost-alternative site (Shabman and Batie, p. 8,
1979).

The concepts discussed above are implicit in Figure 6-1. Recall
that the valuation of wetland sites requires that substitutes be con-

sidered. An ideal study suggested by Shabman and Batie (1981, p. 10)

I. UNI“ It! denounce mum- lo name: utl‘lp

m Sou-zen fess .q .9191) nan-09'1“ mans-no: (smiles-1351 uluamaq

    

bin-1.35m Jana Try-1'1 p.1i'-.:!'”-'Ie '1'. aux." . -"- 1m:- 3-11 i‘T'r-d'seac'c

' .r . ._. . - ' -
l .hu'.‘ ' E ‘-- "f. l'.. '.‘I I. ‘ '- d ..l.-' " ': C"

 

109

”would value the economic surpluses from a wetlands service flow and
compare this value with the cost of providing services by a wetlands
substitute.”
Applications of Methods to Value Freshwater Wetlands
in Improving Water Quality
This section describes attempts to value the water quality services
provided by freshwater wetlands. Several cases for midwestern states

are presented and then evaluated.

Value Studies of Water Quality Effects. Mitch (1979) assessed the

 

value of the Momence wetlands for sediment control and water quality main-
tenance in the Kankakee River in northern Illinois. The cost of replace-
ment of the services of the wetland was the basis for this valuation
attempt. The calculated annual cost to replace these two services for

the 1899 acre wooded wetland was $157,000 or about $83 per acre.

Tilton, et al., (1978) conducted a general value assessment of Michi—
gan wetlands in improving water quality. The researchers assessed the
value of wetlands in nutrient removal (in a wastewater application situa-
tion) and in land runoff removal under two alternate replacement possi—
bilities-—purchase of a wetland and construction of a wetland. The basis
for analysis was that in a developed state a wetland was no longer avail-
able for water renovation. Therefore, the authors state that the
function must be replaced and conclude that ”the value of the wetland
acre is the present worth of the replacement project, over and above the
corresponding wetland project, that would provide the same service to

society.”

   

110

In discussing nutrient removal, Tilton, et al. (1978, p. 74), note
that nutrient removal from secondarily treated wastewater is not a
natural wetland function and thus value of the wetland for this purpose
must be compared with other alternatives. The alternatives include
upland spray irrigation and chemical treatment. Results are shown with
runoff nutrient control in Table 6-2.

Tilton, et al. (1978, p. 81), extraplote their analysis from the
wastewater situation to the removal of nutrients in land runoff. If a
wetland is developed, the runoff nutrient control function is lost. The
authors state the value of a wetland consists of treatment and collec-
tion function since water must be rerouted to an alternative site where
nutrient treatment can occur. Tilton, et al., remark ”that the value of
the wetland acre for nutrient removal from runoff should be comparable

for that from wastewater."

Evaluation of Value Studies. The discussion that follows will be

 

limited to the economic analysis used in previous valuation studies.
Both studies presented utilized the ”replacement cost” or, more famil—
iarly, the alternate cost method. The conditions necessary for proper
application of this method were stated earlier. The fulfillment of the
three conditions will be examined in each study.

The Mitch study poses the question ”What costs would man incur to
perform with his technology the services provided by the Momence
Wetlands?”. The calculation of these costs are given as the value of the
wetlands. This calculation implicitly assumes that society would demand
and incur the costs of the alternatives used to replace the wetlands

services. No documentation is offered to support the assumption that

 

111

Table 6-2

Water Quality Values for Michigan Coastal Wetlands

 

 

Coastal Wetland . Cost of Natural Cost of Net
Function Wetland Function Replacement Value
to be Replaced $/acre $/acre $/acre

 

Nutrient Removal

Purchased Wetland 11,200 (base case)

Constructed Wetland 11,200 25,400 14,200
Upland Spray Irrigation 11,200 20,300 9,100
Chemical Treatment 11,200 21,200 10,000

Runoff Nutrient Control

Purchased Wetland 0 20,400 20,400
Constructed Wetland 0 34,600 34,600
Upland Spray Irrigation 0 29,500 29,500
Chemical Treatment 0 30,400 30,400

 

 

Source: Tilton, et al., 1978, p. 85.

society would in fact demand the services. The condition of the alter—
native replacement project being the least cost project is not mentioned
in the analysis. Finally, the condition that the alternative provides
the same services as the wetland appears to be met although not explicitly
mentioned.

The nutrient removal part of the Tilton, et al., study assesses the
value of wetlands when used for secondary municipal wastewater treatment.
If wetlands are no longer available for wastewater treatment, the alter-

nativesidentified by the authors to provide this function were a

 

 

 

 

 

 

 

112

constructed wetland, upland spray irrigation, and chemical treatment. No
mention of the least cost alternative is made although Table 6-2 indi-
cates upland spray irrigation is least cost. The condition of whether
the alternatives provide the same services as the wetland is adequately
addressed. The major pitfall, however, is in the statement that the
function of nutrient removal must be replaced if a wetland is converted
to another use. Therefore, the authors contend that the value of the
wetland is the cost of the replacement project that provides the same
services. The authors' position is one of implicitly assuming that the
demand would exist for the alternative, even at its higher cost. No
evidence is presented to substantiate this claim. This approach is
extended to runoff nutrient control.3 The inability of this study to
properly apply the alternate cost method casts serious doubts on the
values obtained.

There is much discussion in the literature of the misuse of the
alternative cost technique. The Water Resources Council's Principles
and Standards (1973, pp. 39-40) advocates "cautious” use of the method
only when the willingness to pay or intermediate good method cannot be
used. Further, the guidelines state:

The cost of the most likely alternative means will generally

mistake the total value of output of a plan. This is because

it merely indicates that society must pay the next most

likely alternative to secure the output, rather than estimat—

ing the real value of the output of a plan to the users. This

assumes, of course, that society would in fact undertake the
alternate means.

 

3Note in Table 6-2 the zero costs indicated for the costs of natural
function for runoff nutrient control. There are cost of ownership (e.g.,
taxes) associated with providing the natural function.

'r W
oh a nah-flu me 94*: chive-m mlfltfilf'"

 

1 ed: Jeni anamorsaa ed: n: a? ,19v9woe . feiJiq ucfom ant .bgzagwbb.

b9315vr33 a. nnalruw - l? noseiuei ed - u '" Lrw --4-wrw; iv not soul

:I'! - nil .: - ' 7* . . -. t"“,":‘-L

 

 

113

The guideline notes that some difficulty may exist in determining if the
alternative means would be undertaken. Schmid (1980A,pp. 150-151) notes
in using the alternate cost method, the problem is to establish that
effective demand for the alternate project exists. The essence of the
problem involves estimating the quantity demanded of a particular service
at an increased price. If the price is outside of past observations,
there is none or little evidence to suggest that effective demand exists.

Shabman and Batie (1981, p. 11) note an important point when they
state that when the cost of a substitute is used to measure value of a
wetland, it represents the maximum possible value for the wetlands.

In addition to the problems of the use of alternate cost methods in
the economic analysis, there are possible errors in the value estimates
due to insufficient quantitive data on water quality effects of fresh-
water wetlands. The earlier part of this chapter discussed these data

limitations.

The Potential for Sound Economic Values

In light of the material reviewed in this chapter, it may be useful
to step back and attempt to answer the question: "Is economic valuation
of the impact of freshwater wetlands on water quality possible?" The
answer to this question is not a simple yes or no. Some knowledge of
the processes occurring in freshwater wetlands affecting water quality
exists. Much about the processes is also unknown. Little quantitative
data in complete input—output mass-balance form exist for freshwater
wetlands. Hydrologic regimes are important factors in water quality

interactions and have not been intensively studied. Studies from other

ins um“?- Ill:
.Sfi-ffl rate-w sun-Hits on: 10! bum w

 

solvwaa'ialmrfie‘q s to bean-met. (.1. .r'suv 4.1: m'vie'c'rtae eevfpvni mama
1 .3"-"'".'It.‘-r"l'.1."ul.‘ 7264- ‘1 '--' 1.1-: a" '< I-. - ' "" -- --' ‘. era-«.21? 116 II

..III I I- I “11:" I 'I-ca.“.'.l . I I _ fl'lfiflj

114

geographic areas and from coastal areas indicate possible water quality
benefits from wetlands. Foster (1979, p. 85) notes that wetland bene-
fits vary by type of wetland and from place to place. Burton (1981,

p. 6) further points to the differences in wetland benefits across geo-
graphic areas with respect to water quality effects of freshwater river-
ine marshes by emphasizing the risk of extrapolating results of studies
in Wisconsin to other midwestern states.

The economic valuation techniques appear to be available to measure
the water quality effects of wetlands. Up to now, however, many value
studies have improperly utilized the alternate cost method probably
resulting in an overstatement of wetland water quality benefits.

Application of the economic tools has some practical limitations.
All the valuation methods depend on accurate understanding of the physi-
cal production processes involved. Improved understanding and quantifi-
cation of the technical linkages between wetlands and wetland services
is needed to value services such as water quality improvement (Shabman
and Batie, p. 13, 1981).

The variability in productivity of wetlands creates the need for
detailed biological assessment of the production processes of specific
sites in order to determine their economic value. These studies are
expensive but necessary due to the site—specific demand for wetland ser-
vices and variable productivity. General value estimates cannot be
applied to specific wetland sites (Shabman and Batie, p. 10, 1981). In
summary, the accurate estimation of natural wetland values will require

expensive and time-consuming data collection on individual wetland sites.

 

CHAPTER VII
SUMMARY AND CONCLUSIONS

The final chapter of this thesis presents a summary of the important
issues and findings, offers conclusions for wetland resource management,

and suggests additional research.

Summary

The purpose of this research effort was to assist and improve
choices about the use of wetland resources in general and Michigan wet-
lands in particular. The first step in the investigation was the devel—
opmentcfliabroad conceptual framework for visualizing and analyzing
natural resource problems. This framework was applied to the wetland
resource and it has proved useful in understanding the nature of the
wetland allocation problem. Important elements in the framework are the
notions of interdependence, external effects, ownership and property
rights, uncertainty of impacts, scientific industrialization as a driv-
ing force, protective institutions, and the dynamic or evolutionary
processes involved. Application of the framework provided insight into
the nature of man's interaction with the wetland resource, how this
interaction produced conflict over access to alternative wetland ser-
vices, and the resultant formation of wetland protection laws.

The second segment of the investigation began with the assumption

that wetland protection was a valid public purpose and addressed the

115

 

 

116

allocative questions of ”How much wetland protection?” and ”Which
wetlands are to be protected?". These questions were addressed in the
context of the Michigan permit decision system which requires the bal—
ancing of benefits and costs of proposed activities on wetlands. Funda-
mental concepts from the discipline of economics were used to construct
an approach to systematically consider the crucial elements in wetland
permit decisions. The approach emphasizes the need for information on
the societal value of both preservation and development wetland uses

for effective policy choices.

The investigation's third part attempted to fulfill the informa-
tional requirements of a wetland permit decision by estimating preser-
vation and development values of wetlands in southeastern Michigan. A
methodology for estimating the developmert value ofyetlands for waterfront
residential property was developed and applied. The methodology's
strengths, weaknesses and limitations were assessed.

The economic value of freshwater wetland impact on water quality
was investigated in a review of the literature. No valuation of this
natural service was undertaken because of lack of conclusive quantita-
tive data on the relationship between wetlands and water quality in

Michigan.
Conclusions

The conclusions are addressed in four categories: the wetland
resource allocation decision, methodological issues, implications for

Michigan wetlands, and the role of science in natural resource policy.

 

 

117

The Wetland Resource Allocation Decision

 

The Goemaere-Anderson Wetland Protection Act of 1979 recognized the
functions and values of Michigan wetlands and created a mechanism to
ensure wetlands are protected and preserved. The allocative question is
how many and which wetlands are worth preserving. The Goemaere-Anderson
Act consists of the guideline of permitting wetland conversion whenever
a balancing of benefits and costs indicates the project is in the “public
interest." This guideline as well as the basic economic concepts of
opportunity cost and marginality suggest that the economic values of
wetlands in preservation and development uses are essential for effective

wetland permit decisions.

Development Value. The conclusions resulting from the estimation
of the development value of wetlands for waterfront residential use are
as follows:

(1) The empirical results of the study indicate that wetlands for
which no alternative development site exists can have substantial devel-
opment values. When an alternative development site was identified,
development values were considerably lower. In general, to the extent
that alternative substitute sites exist to the development of wetlands,
development value is reduced.

(2) Development value is highly dependent upon location meaning
that estimates are highly site-specific and should not be extrapolated
to wetlands in other areas.

(3) The methodology and application focused upon the estimation of
small (marginal) additions to the supply of waterfront land by develop-

ing wetlands. Therefore, the values found here for specific sites

 

118

should not be expected for all area wetlands in trying to estimate

overall development value of a large wetland area.

. 1.:
9957:

J Preservation Value. The preservation or natural value of wetlands'
effect on water quality was also addressed in this study. Conclusions
regarding the value of this wetland function are below.

(1) A review of the literature on the effect of Michigan freshwater
wetlands on water quality revealed a lack of conclusive quantitative
data on this relationship. This is not to say that the relationship
does not exist, but rather that it is not well understood at present.
Extrapolation from studies of other regions and wetland types was not
attempted due to possible erroneous conclusions. Therefore, estimation
of the economic value of this wetland function was not undertaken.

(2) Some general statements about preservation values of wetlands
can be made based on the literature review done here and other authors
work (See Foster, 1979; Shabman and Batie, 1981). As expected, wetland
preservation values vary from one location to another due to differences
in productivity and the demand for wetland services. Thus, preserva-
tion values are also site-specific and cannot be generalized or extra—

mflatedto wetlands in other locations.

Methodological Issues

 

Development benefit estimation was accomplished through use of the
hedonic price technique which has some limitations. As used here the
technique was only capable of estimating small increases in the amount
of waterfront land created from wetland development. Case study areas

appeared to comply with the limiting assumptions of the hedonic method.

__ ..- ‘i - .'--= 1,... r91:-

magma-11:121.,

blzllnhbo ozla 26w vailaup 19:5u no 336?}.

    

.on9d 916 notionu? bnrfjhw aid- 1“ eelsv 9d: gnib1595fi

‘1':-'::'- - -'.'.—.!-12-."1 9:3 .- . ' 7"-

'dsiswdeswfi nspiduir to

'1. -.: -_1. ‘.!:1.-.: -.- 2.15 --‘1.'_

119

The most significant problem encountered in the application of the
hedonic price method was the lack of appropriate substitute sites in
three of four case study areas. The assumption made for these three
areas was that no alternative site existed which could provide the same
services (waterfront location amenities). If this assumption is incor—
rect, then development value estimates are upwardly biased. A possible
substitute site was identified for the remaining study area. As
expected, estimates were lower when the substitute site was considered.
In summary, the application indicated that careful measurement with
full consideration of alternative sites is a prerequisite to estimating
sound economic values.

The review of preservation value studies of the wetland-water
quality relationship revealed that the alternate cost method has been
inappropriately used. Also, since preservation values are site-specific,
region or state-wide value estimates are suspect.

The concepts of alternatives and substitutes are essential to esti-
mation of preservation values. The preservation value of a wetland is
reduced to the extent that alternatives or substitutes exist to the
provision of wetland services by the preservation of wetlands. Thus,
careful systematic analysis using a common set of economic concepts is
required to produce sound and accurate estimates of economic value of

either development or preservation uses of wetlands.

Implications for Michigpn Wetland Resource Poliqy

 

The objective of this section is to present the possible implica-
tions of the results of this study for Michigan wetland resource policy.

The discussion of implications will proceed at two levels. First, the

 

119

The most significant problem encountered in the application of the
hedonic price method was the lack of appropriate substitute sites in
three of four case study areas. The assumption made for these three
areas was that no alternative site existed which could provide the same
services (waterfront location amenities). If this assumption is incor-
rect, then development value estimates are upwardly biased. A possible
substitute site was identified for the remaining study area. As
expected, estimates were lower when the substitute site was considered.
In summary, the application indicated that careful measurement with
full consideration of alternative sites is a prerequisite to estimating
sound economic values.

The review of preservation value studies of the wetland-water
quality relationship revealed that the alternate cost method has been
inappropriately used. Also, since preservation values are site-specific,
region or state—wide value estimates are suspect.

The concepts of alternatives and substitutes are essential to esti-
mation of preservation values. The preservation value of a wetland is
reduced to the extent that alternatives or substitutes exist to the
provision of wetland services by the preservation of wetlands. Thus,
careful systematic analysis using a common set of economic concepts is
required to produce sound and accurate estimates of economic value of

either development or preservation uses of wetlands.

Implications for Michigan Wetland Resource Poligy

 

The objective of this section is to present the possible implicae
tions of the results of this study for Michigan wetland resource policy.

The discussion of implications will proceed at two levels. First, the

.‘u’m were-e
alt: outrin1ssf| on SIN! tiw'IIIII

 

'il .(zgtrfnsma nnrssoof :nowliaaaw) aaotv1le

' HTUTZIOI] A .beaal'i vlb'ilav‘d 9' C 293.11" '- 11:,-1-'m:*!;--.-;-'1 :15!” .3351

3| i.
'-.-.~'- .-'--_"-5 1151.52 r'n'. --". -.- -1' -. '- "':'r:'. 'Eii'. . ..I" '-':‘uz

120

implication of conclusions about wetland values will be examined in a
general policy setting in which the institutional mechanisms for wetland
protection are a variable. Secondly, the implications of study results
for the existing policy setting of the permit process of the Goemaere-

Anderson Wetland Protection Act will be presented.

General Policy Implications. As noted, both development and pre-
servation values of wetlands are expected to be site-specific, making
generalizations about relative magnitudes of these values difficult, if
not impossible, to make. Two implications of this result are:

(1) A policy instrument or mechanism which takes account of indi-
vidual site characteristics is preferable to one that ignores site char-
acteristics. For example, the individual case by case approach to
wetland use decisions is preferable to any large-scale preservation or
development decision that ignores individual characteristics. The per-
mit process of the Goermaere-Anderson Wetland Protection Act is consist—
ent with this conclusion. However, a shortcoming of the Goemaere-
Anderson Act may be the categorical exemption (large-scale decision) of
several wetland uses. Exemption of these uses implicitly presumes that
the development benefits of such uses always exceed costs. This may
not always be true due to differences in site characteristics and demand
that affect wetland preservation and development values.

(2) The inapplicability of general natural value estimates to
specific wetland sites means that detailed biological assessment is
required for valuation of such sites. For numerous small wetland
areas, careful economic valuation of natural wetland will

entail much time and effort. Similarly, development value

 

121

estimation must be done on a site-specific basis. As noted, accurate
estimation of either development or natural values necessitates careful
systematic analysis, requiring much time and information. These costs nay
significantly add to decision making costs under a case by case permit
system. These costs are one dimension of performance of the permit
approach as a policy instrument for protection of wetlands. The com-
parison of the permit approach with other instruments for wetland pro—
tection (taxes, subsidies, easements, or transfer of development rights)
with respect to this and other performance dimensions would improve the

ability to effectively and efficiently manage wetlands.

Implications for Implementing Michigan Legislation. Several prac—

tical implications of the study's results for the implementation of the
Goermaere—Anderson Wetland Protection Act are discussed below:

(1) As noted, the permit process should proceed on an individual
case by case approach to consider site-specific characteristics of
wetlands. Large-scale permit decisions are not recommended.

(2) The site-specific estimates of development value found in this
study can add to the store of knowledge with defined attributes. These
cases can be considered observations leading toward more defensible
generalizations.

(3) The empirical results of studies such as this can assist in
compiling a checklist of factors that influence development and preser—
vation values of wetlands. Important factors determining development
value are location (accessibility) and availability of substitutes. The
major factors influencing preservation value are productivity, location

(demand for services), and the availability of substitute sites. Future

 

    

MI! Mulch

mamb o: m gunman-mk-

“To aanllndifiia to naiaannib an; 915 arena sasdT .nliltc

-mpa sdT .abnsfisw io nai:u“*n~fi w ’ - - t~wr uziiru n u- dshafiqoa

'm- Hm.

'1 1' :l '(l ‘-

"Hal 1: ..r m g. . n . I. I .l. .. .. - . -'.I ' ...,‘n

122

studies could expand this checklist. One possibility is that factors on
the checklist could be utilized as a decision making aid. For example,
field workers could assess the contribution of important value determin-
ing factors on a scaled basis for individual wetland sites in order to
compare preservation and development benefits. This procedure would not
substitute for economic valuation but could circumvent the high cost of
case by case valuation attempts while still maintaining the spirit of
systematic analysis.

(4) Another practical suggestion may be the use of an empirically
derived checklist as described above to screen the "easy” permit deci-
sions from the more difficult ones. In other words, whenever the check-
list system produced unclear or indeterminate results, more careful
economic valuation could be concentrated on these difficult decisions.
The procedures and concepts in Chapters IV, V, and VI are essential for

economic valuation attempts in these decisions.

The Role of Science in Natural Resource Policy

 

The final topic to be addressed is that of the role of science in
natural resource policy. Science clearly has a role to play in resource
policy, and that role is demonstrated very well in the evolution of
wetlands policy in the United States. Scientific information about the
impacts and consequences of the destruction of wetlands created the
awareness of wetland functions and potential values. As a result, many
laws have been passed to protect wetlands. This illustrates the role
of scientific information as an input into the political process of
choice. The choices involve who and how different groups will benefit

from resource use.

U“. 1.. .
M thud bur-n: o no mini ii
in IPiN m 2??" .atisned inmohvsb an: noimvwszswq 91m:

    

to no: drain a!!! ”swam-('1 Man-- “J- 1...: in ”means .~'--.'*'r32du:

._ an? I.- I" . .."- _ _-, .. ' . ._ ._..53

 

123

The form that much of wetland protection policy has taken heavily
emphasizes physical and biological information in management and decision—
making. This is mainly because of the biological functions on which
wetland protection laws are based. Walker (1973, p. 92) comments on
scientific management in the following way: "The basic flaw in scien-
tific management is that it is founded on an understanding of the natural
system and not the human system.” Walker further states that ”an
obsession with the exploration of physical processes obscures the more
important task of understanding and controlling the social processes
that lead man to alter nature.” He concludes that natural resource
management requires the management of social and economic systems as
well. Scientific knowledge alone cannot be transformed into effective
resource policy. One nagging problem in wetlands management is the
limited understanding of complex ecosystems and the inability to accu-
rately predict impacts of man's activities upon them.

The role of social sciences, especially economics, has been stressed
throughout the thesis. Many calls have been made for economists to come
forth and value wetlands. Economic attempts at valuation of wetland
natural services have generally not been that successful. Still, the
position taken here is that economists do have a role in formulation and
implementation of wetland policy choices. The role is to help society
to properly conceptualize a problem and to ask the right questions
(Bishop, 1979, p. 379). The role involves the identification and pre-
diction of alternative consequences of particular choices. The purpose
is to assist choice, not to replace it, by providing information on the
consequences of alternative choices. The essence of this thesis was to

do this by delineating the consequences of the choice of either

 

lawn." 9. _ Iflfijififi m .1_..--_.

124

preserving or developing a wetland. The overall conclusion is that
economic information, as well as physical and biological information,
can be a useful input in the political choice as to what use wetland
resources are put.

In conclusion, political choices about rights to wetland use are
inevitable. The question is one of collective social choice as to what
wetlands are to produce (public goods or private goods). The question
is not simply answered by technical expertise or an economic decision
rule. The role of science (social, physical and biological) in natural
resource policy is to provide accurate, timely, and sufficient infor-

mation for effective choice.

Recommendations for Future Research

 

There are many possibilities for future research that would improve
our knowledge of the value of wetlands and assist in policy decisions
concerning their use. Several possibilities are presented below.

(1) The further exploration of the biological and physical produc-
tion functions of wetlands is an important research area if informed
wetland use decisions are to take place. We need better data on what
types of services are generated by wetlands and the quantity of those
services provided under varying conditions. Biological and physical
scientists will play a central role in pursuing this research area.

(2) The economic estimation of natural wetland values is another
fruitful research area,provided more quantitative knowledge about wet-
land functions is obtained. The careful economic analysis required for
sound value estimation should be carried out in cooperation with econ-

omists or in a multidisciplinary mode.

125

(3) Further work by economists is needed to refine and improve
existing value estimation techniques as well as to develop new tech-
niques. The hedonic price technique used in this study is just one
method having potential for greater use and refinement.

(4) Another area of inquiry is the investigation of the factors
that led wetlands to be altered. This inquiry could pursue further ana-
lysis of the political-economic dimensions of wetland resource use.

The role and motivations of the important actors (scientists, bureau-
crats, legislators, associations, and interest groups) in the political
economy could be empirically studied.

(5) The relative efficiency of alternative institutional mechan-
isms (policy instruments) for protection of wetlands is an extremely
important research area. The performance dimension of decision making
cost is just one criterion for institutional comparison. Alternative
mechanisms to the permit system for wetland protection available for
study include taxes, subsidies, easements, reassignment of rights, and
transfer of development rights mechanisms. Effective policy requires

choice among institutional alternatives based on evidence of performance.

 

APPENDICES

 

 

APPENDIX A
COMON PROPERTY OWNERSHIP AND BEHAVIOR

126
Appendix A: Common Property Ownership and Behavior

The behavior of individuals in common property ownership situations
will be discussed in the context of the common property resource of the
fishery. Figure A—1 illustrates a production function in a static model
of the fishery. The model is static or ”steady state“ because the fish
stock is maintained at a constant level. Catch rate (Rt) and fishing
effort are plotted along the vertical and horizontal axis, respectively.
The total cost (TC) curve, which represents the cost of capital-labor
inputs into fishing effort, can be constructed if w can be assumed to be
the cost of one unit of effort (Howe, p. 263, 1979). If the prices of
fishing effort inputs are unaffected by the level of fishing effort, then
total costs rise at a constant rate. The marginal and average cost
curves can be derived from the TC curve and are shown in the same hori-
zontal line in Figure A-2 (Nelson, 1979; Gordon, pp. 133-134, 1954),

The production function or total product (TP) curve in Figure A-l
demonstrates the ”law of diminishing returns" to fishing effort. As
fishing effort increases, the catch rate increases at a decreasing rate
until it begins to decrease. The marginal and average product curves
are derived in Figure A-2 and reflect diminishing returns in their down-
ward slope. The marginal product (MP) curve indicates that the marginal
t is
the level of effort required to reach the maximum substainable yield.

productivity of fishing effort declines and goes to zero at Lt' L

Unless the capital-labor inputs to fishing effort are available at zero
cost, it does not pay to increase the catch rate to the maximum substain-

able yield (Howe, p. 262, 1979).

Catch Rate R

)
\

1

Catch Rate

 

127

 

 

 

 

 

 

 

Figure A—l
Fishery Productivity Relationships
\" /, ’ TC = L * w
/ /
/ /
x
/ // I
/ / I
/ TR = TP * P
.f.‘
l
‘ l
l . . L
L* L L
t t t Fishing Effort L ‘
(t)
.Figure A—2
Average and Marginal Relationships in the Fishery
E
-d... _ _.__ -.. u l
l\
d C \l AC, MC
‘ I\AP or AR
1
I ,
l e
) ‘ 1L t
L*t Lt HP or MR

Fisning Lffort L.
(t)

 

128

A second important result can be obtained from Figure A—1 and A—2
if the price of fish can be assumed to be equal to one. The TP curve
becomes the total revenue (TR) curve representing the total value of all
fish caught. The MP and AP curves become the marginal revenue (MR)
and average revenue (AR) curves. The difference between the TR and TC
curves in Figure A-1 represents the value (benefits) of the fishery
resource to society (or the profits accruing to fisherman). These net
benefits are maximized where the difference between the TR and TC
curves are at its greatest point which is at effort level L*£. The same
level of effort is found in Figure A-2 where the MR and MC curves (slopes
of TR and TC curves) intersect. The net profit generated at effort level

L*

t is XY (Howe,p. 263, 1979).

The behavior of individual fishermen under common property ownership
can be explored using Figure A-1 and A-2. Profit XY is a pure economic
profit or rent meaning that it is above what is required to keep the
capital-labor inputs (used in fishing effort) in the industry. In other
words, the capital and labor inputs are earning more than their oppor-
tunity costs. The profit or rent can also be illustrated using Figure
A-2. The area abcd represents the rent yielded by the fishery resource.
L*t is the optimal rate of use of the fishery and the rent reflects the
productivity of the particular fishing area. The critical theoretical
point is that since the fishery resource is not private property, its
rent is not capable of being appropriated by anyone. Any individual can
fish in any particular area he or she desires (Gordon, pp. 135-136, 1954)

The existence of profit or rent attracts other fishermen to the
industry. If free entry exists, as it does in the common property own-

ership case, capital-labor inputs will increase to Let, exhausting all

 

 

129

the net profits or benefits in the industry (Howe, p. 263, 1979).
Potential entrants presume that it is possible to earn the average re-
turn existing in the industry. As a result, capital-labor inputs enter
into the industry up to the point where the average return from fishing
(AP or AR) just equals the cost of fishing effort (MC or AC). The
resulting fishing effort level is Let. At this point, the MP or MR of
fishing effort is negative and the fishing level exceeds the maximum
sustainable yield. The reason this occurs is that no single individual
has the incentive to take account of his or her impact upon productivity
(Nelson, 1979). Instead, each person cares only about the average pro—
ductivity or return from fishing rather than the marginal productivity
or return (Gordon, p. 136, 1954). The important point of this discussion
is that if there is free access to a common property resource, then
excessive capital-labor inputs will flow into the industry, resulting

in an excessive reduction in the stock of the resource (Howe, p. 263,
1979). Thus, in many cases, free entry and individual competition can
result in degradation, overexploitation, and possibly destruction of a
resource. In a sense, the statement "everybody's property is nobody's
property“ has relevance for common property resources. Gordon (p. 141,
1954) summarizes the problem ”wealth that is free for all is valued by
no one because he who is foolhardy enough to wait for its proper time

of use will only find that it has been taken by another.“ Thus, Gordon
notes that common property resources are free goods for individuals and
scarce for society. Under common ownership, natural resources can yield
no rent because of unregulated private use. Conversion to private or

public property is needed for a rent to be yielded.

130

There are situations where common property resource ownership does
not result in overuse and depletion. Consider the case where a resource
is not in great demand relative to supply. This situation is illustrated
in Figure A-3. Using the fishery example, suppose the price of fish is
decreased to $.50 which reflects the interaction of supply and demand.
The TR curve drops to TR2 curve. The impact of this movement upon
resource use is that the free or open access level of fishery use de-

. e .
clines to L t

which is less than the maximum sustainable yield. Thus,
common property ownership in this case does not result in depletion of
the resource. The implication of this example is that common property
ownership can work in one region and not another depending on the rela-
tive supply and demand for the resource in question. More specifically,
residents in unpopulated regions (such as northern Michigan) may have
very different attitudes toward natural resource management issues than
residents in populated regions (southern Michigan). This creates a
problem of what the relevant jurisdictional boundaries are to be in

management policies which is further complicated by spillover effects

from one region to another.

 

Catcn Rate R

Catcn Hate R}.

(t)

it)

 

131

Figure A-3
Fishery Total Revenue and Cost Relationships

TC

TR
l l 3
Le- .3.
t Fisning Effort L(
t

Figure A-4
Average and Marginal vacnue Curves in the Fishery

 

 

 

f
I\

 

‘ AC, MC

 

>
I}

 

I
Let \\\\\ Fishing Effort L
|.l< (t)

 

 

APPENDIX B

PROPERTY VALUES AND BENEFIT MEASUREMENT

 

132

Appendix B: Property Values and Benefit Estimation

It is evident that the productivity of land parcels differs among
sites. Classical rent theory suggests that productivity differences of
land produce differential land rents and, therefore, differential land
values. Whenever the productivity of a land parcel is greater than its
rent, the activity located on that land is earning an additional or
surplus profit. For a producer's good, competition and free entry are
sufficient conditions to guarantee that surpluses and profits will be
eliminated. This is because individuals interested in the more produc-
tive land are free to enter and bid above existing land rents to secure
location on that land parcel. The bidding process eliminates the surplus
for the new buyer, thus ensuring that land values fully represent rent
and productivity differences (Freeman, p. 108, 1979).

When land is used primarily as a consumer good the above result may
not hold. It is possible that some consumers or households are occupy-
ing land at a rent that is less than their total willingness to pay, but
that other households are not interested in outbidding them. The result
is that an economic surplus is accruing to the household at this more
productive location. In such a case, land values may not fully reflect

rent and productivity differences (Freeman, p. 108, 1979).

Use of Land Values to Measure Benefits of Environmental Improvements

 

Environmental characteristics of land such as air quality, noise,
and proximity to water can also affect a land parcel's productivity.
Thus, environmentally determined productivity differences will be re-
flected in the land rent structure and ultimately should appear in land

values as long as no surpluses exist. This result has attracted much

133

attention from economists interested in measuring the value of output
from public projects that affect the environment. Because of the public
good (non-excludable and non—rivalness) nature of many environmental
improvements, their value is uncertain. However, it is hypothesized
that individuals choose their desired quantities of local public goods,
such as air quality, in their choice of residential location. If indi-
viduals have freedom of choice, it is possible that information on the
demand for and value of environmental improvements can be inferred from
prices and quantities eminating from private housing markets (Freeman,
p. 78, 1979).

Another explanation of the rationale for observing land values re-
lies on the concept of a factor controlling access to a publicly provided
good. For example, in order to benefit from a public project improving
air quality, an individual must reside in a location where the benefits
of air quality improvement occur. The cost of access to these locations
is bid up as individuals attempt to gain access to the improved area.
This produces a rent to parcels in the improved area. The total value
or benefits of the air quality improvement can be computed by the addi-
tion of all these rents. However, if any surpluses are present, the
summation of the rents understates the value of the project. A typical
approach used to measure the change in rents is to compare differences
in land values of land with access to a project's benefits and similar
land which has no access to the project's benefits (Schmid, p. 143,

1980A).

 

 

134

Property Value Changes and Rent Theory

Property value changes are defined as the present value of antici—
pated changes in the flow of rents from a parcel of land. The benefits
of an environmental improvement has been defined as the total monetary
value placed on the environmental improvement by all persons directly or
indirectly affected by the improvement. The dollar valuations of these
persons is termed their "willingness to pay” for the improvement (Free-
man, p. 3, 1979).

The main concept of classical rent theory relevant to the matter of
using property values changes as benefit measures is that any utility
generated by a public project appears in economic rent so that utility
or non—land factor return is equalized on all qualities of land when
individuals try to obtain access to the project through bidding (Schmid,
p. 144, 1980A). As mentioned above, property value changes only accurately
measure benefits when all potential surpluses are eliminated by the bid—
ding up of rents and property values of an environmentally enhanced site
by interested users. When this has occurred, all potential users should
be indifferent to improved sites with higher rents and unimproved sites
with lower rents. This means that a land owner's rent is exactly equal
to the willingness to pay of all potential land users. It should now
be evident that only when no surpluses exist do rent and property value
changes represent project benefits (Freeman, p. 112, 1979).

With the importance of the absence of economic surpluses in mind,
let us examine the important factors that determine whether surpluses
are eliminated. The first factor is the pattern and determinants of the
demand for land. If land is used as an imput into a production process,

the demand for land is a derived demand that depends on the demand for

 

 

135

the output and the production function involved. For example, say land
is used in the production of corn. Assume a region has homogenous land
except for the distinction that land in the western section is exposed

to air pollution. The pollution reduces the productivity of land for
corn production. This causes lower rents in the west in the resultant
equilibrium. The factor forcing rents down in the west is that producers
returns net of rent must be equal throughout the region. This process

is termed factor price equalization (Freeman, p. 112, 1979).

Suppose a public project is initiated that eliminates air pollution
in the west. The productivity of all land within the region is now equaL
If the increased output from the western land causes the price of corn
to drop in the market, some of the benefits of the project are received
as surpluses to corn consumers. Such benefits are not reflected in land
rents and, thus, will not be observed in property value changes. This
result indicates that a necessary condition for land rent and property
value changes to accurately measure benefits is that there be no changes
in the price of output or prices of other inputs. If these prices remain
constant, land rents in the east are unaffected by the environmental
improvement in the west. Western land is now equivalent to eastern land
in its productivity for corn. Thus, rents must be identical in both
areas. The rise of the rents in the improved area correctly measures
the benefits of the public project that improved air quality (Freeman,

p. 112, 1979).

The second important factor determining whether surpluses are eli-

minated is the nature of the land market. The important question is

whether the land market under consideration is part of a larger open

136

economic system. Consider a region similar to the above, the only
difference being that there are many types of activities located on
sites within the region. Suppose that air pollution is eliminated as
before in the west. The initial equilibrium is disturbed, causing some
activities to relocate within the region and some activities to leave

or enter the region. These relocation effects have been discussed by
several authors (Freeman, p. 113, 1979). The first was Strotz who
argued that benefits should be measured as the sum of the absolute
values of the changes in rents of the improved and unimproved sites. To
illustrate, suppose a town has an air pollution problem that affects all
parts equally. The town is divided into two halves, north and south. A
public project improves air quality in the north and consequently north
land rents are bid up while south land rents are bid downward. Strotz
contends that the decline in rents in the south is as much a benefit

as the gain in rents in the north. For this reason, he suggests that

an accurate measure of benefits is the sum of the absolute values of the
changes in land rents (Strotz, p. 176, 1968). This paradoxical result
can be clarified by noting that Strotz defined benefits for the two
areas assymetrically. In the north, Strotz terms the increase in land
rents as benefits while in the south he defines benefits as the increased
surpluses of consumers enjoying lowered rents.

Lind (p. 201, 1973) has maintained that he has established two
important contributions on the matter of accounting for relocation ef-
fects. First, he demonstrated that the benefit of an environmental
improvement can be measured as the change in profits or surpluses of
activities that locate on land parcels directly affected by the land

improvement project. He notes that although rents and surpluses of

 

'. Mfg-H.-
- III

 

 

137

activities on unimproved parcels may change most of these effects cancel
out. Secondly, Lind concludes that the net change in value of directly
affected land is an upper bound approximation of the benefits of improve-
ment. However, rents on this land must be created in such a way that
consumer surpluses and profits are eliminated. The implication of this
is that benefits can be estimated as the change in the improved land
alone and the change in values of unimproved sites can be ignored (Lind,
p. 202, 1973).

Freeman, in a comment on the Lind article, emphasized that the use-
fulness of land value changes as an estimate of benefits heavily depends
upon satisfying the condition of zero surpluses. Also, the ceteris
paribus assumption (all market prices except land rents are unaffected)
must hold. Freeman (p. 472, 1975) points out that satisfying the two
assumptions simultaneously is not likely to happen in practice. Thus,
it is not likely that land value changes are useful as an estimate of
the benefits of environmental improvements because of the restrictive
nature of the needed assumptions.

A closer analysis of the Lind paper reveals some interesting limita-
tions of his conclusions about relocation effects. The case presented
by Lind states that an environmental improvement project causes activi-
ties not located in the improved area to move into the improved area.
The net change in total productivity of all activities relocating from
the old to the new equilibrium is given as D. Lind gives a measurement
for the possible upper and lower bounds of D. The upper bound is repre—
sented by the increase in the profits or surpluses that would be created

by the activities moving onto an improved site if the land rents were

 

 

138

the same before and after the project. The lower bound of D is measured
as the same thing only with respect to the equilibrium land rents after
the project. In other words, the net increase in surpluses of activi—
ties located on improved land after the project is an upper bound of
benefits if measured in initial rents and a lower bound if measured in
final rents. Lind goes on to argue that if the only land rents that
change are those of improved sites and that the activity displaced by
the project is marginal to all unimproved land, then the upper bound
measure of D is an exact benefit measure (Lind, p. 199, 1973). Thus, it
would appear that Lind's argument that measurement of only enhanced land
rent and property values are necessary for benefit estimation depends on
the assumption that displacement of land due to the project is marginal
with regard to the total supply of that land. The previous statement is
further supported by Lind's recognition that his techniques do not take
into account locational interdependence and, therefore, cannot be used
for the evaluation of redevelopment of an entire region.

Polinsky, Shavell, and Rubinfeld have also considered questions
concerning the predictability and interpretation of changes in property
values caused by an improvement in environmental amenities. Their model
assumes identical utility functions and equal incomes of all individuals
in the city where an improvement takes place. The residential location
decision can be stated as:

Max U [x,q,a(k)] subject to Y = x + p(k)q + T(k) Equation (1)
Where:

x = consumption of private good

q = consumption of housing

k = distance from central business district

“1

 

139

a(k) = index of amenities
p(k) = price per unit of housing
T(k) = transportation cost

Y = income

. The indirect utility function V is used in the mode. V expresses the
household's utility as a function of prices at a particular site, net

income (transportation costs netted out), and amenities at that location.
V(k) = V [p(k), Y—T(k), a(k)] Equation (2)

Since individuals can freely move in order to increase their utility,
the equilibrium pattern of property values exists when all individuals
have increased their utility to the maximum possible. In other words,
some equilibrium level of utility V* exists that is independent of loca-

tion. This relationship is shown in Equation 3 below.
v* = v [p(k), Y-T(k), a(k)] Equation (3)
The relationship between property value p(k) and V*, T(k), a(k) is

implicit in Equation (3). This relationship is shown below.
p(k) = f [V*, Y-T(I<), a(k)] Equation (4)

Equation 4 indicates that given the relevant characteristics of land at
each distance for the center of the city an equilibrium property value or
rent function can be found. The most important feature of Equation (4)
is that V* is included as an argument in the equation. The determination
of the equilibrium utility level V* deserves discussion. This will be
done in the context of two extreme versions of the model (Polinsky and

Shavell, p. 103, 1975).

 

140

The first case to be discussed is called the “small open-city"
model. This version of the model assumes that consumers are perfectly
mobile both within and among urban areas. Perfect mobility implies that
movement has no cost and knowledge of alternatives is complete. This
assumption means that there are in effect limitless opportunities else-
where. As a result, there is a level of utility V* that is common to
each class of individuals that can be considered exogenous to any one
area in the equilibrium of a large system of open cities. This means
that an improvement in the environment at one area cannot affect the
endogenous level of utility V* (Polinsky and Rubinfeld, p. 162, 1977).
This result can alternatively be explained by saying that supply and
demand forces throughout the system dominate the supply and demand
forces in the small open city. They so strongly dominate that the sup-
ply of land in the small open city has no effect on the total supply of
land in the system, and, therefore has no effect on the rent earned by
particular parcels in that city (Polinsky and Shavell, p. 123, 1976).

The importance of the fixed exogenous utility level in this model
can be illustrated by observing the effect of an increase in the level
of environmental amenities at a site in a small open city. Equation (3)
shows that if utility must be held constant at level V*, then an upward
shift in the level of amenities will cause an upward shift in the rent
schedule. This can be seen by examining Equation (4). In other words,
the rent at any site is dependent only on the level of amenities at
that site since everything else in Equation (4) is constant.

For example, suppose that the level of environmental amenities is

increased over a section of a small open city. The utility of residents

   

 

141

in that section is temporarily increased. Migration into the city occurs
which drives up land rents causing the utility level of residents to
decrease back to its original level. The city also expands. The rise

in rents just offsets the increase in utility associated with environ-
mental improvement (Freeman, p. 116, 1979). There has been no increase
in willingness to pay of renters but the profits of land owners will have
increased by an amount equal to the change in aggregate land values.
Thus, the change in aggregate property values represents the willingness
to pay of all participants (Polinsky and Shavell, p. 121, 1976).

The second case is called the "closed city" model. In this version
of the model, the land area and population of the city are fixed although
movement within the city is costless. This implies that no migration
into or out of the city is possible and no expansion of the urban area
can occur.

Now consider an increase in the level of environmental amenities in
a section of a closed city. The environmental improvement causes adjust-
ments and relocations to take place that usually result in an increase
in utility for those who reside in the improved area. The improvement
may result in either an increase or a decrease in the equilibrium level
of utility of all residents of the city. The new equilibrium indirect

utility function and rent function are shown on the next page.

V**

V fi3(k)a Y-T(k)9 a(k)] Equation (5)
p(k) = f V**. Y-T<k), a(k)] Equation (6)

From the equations above it is evident that property values will change
after the environmental improvement since both the new utility level V**

and the amenity level at improved sites will be different. V** is

   

 

142

affected by changes in amenities at other sites in the city. The new
equilibrium level of utility V** is now endogenous to rent Equation (6)
because the city is isolated from the rest of the system. This means
that new rents after the environmental improvement cannot be predicted
from the rent equation because a ceteris paribus relationship between
p(k) and a(k) no longer exists. Also, the change in aggregate property
values or land rents no longer represents the willingness to pay of
residents of the city for the specified environmental improvement. The
reason for this is V** is now endogenous in the equation and is affected
by amenity levels at other sites in the city. In order to use Equation
(6) to predict property value changes it is necessary to know the new
level of utility V** (Polinsky and Shavell, p. 103, 1975).

Two important points concerning changes in land values as a measure
of the benefits of an environmental improvement are worthy of restate—
ment. First, property value changes only measure benefits when some
mechanism is at work to eliminate economic surpluses that might result
from the improvement. Lind's assumptions about competitive markets and
perfect mobility are an example of such mechanisms that can be used to
ensure surpluses do not exist. Also, interurban mobility is the mechan—
ism in the Polinsky-Shavell model to ensure that surpluses are eliminated.

Secondly, the elimination of surpluses is necessary but not suffi-
cient to estimate property value changes from econometric rent functions
such as Equation (6). Polinsky and Shavell (1976) identified underlying
parameters of utility functions using an empirical estimation of the
rent function. They utilized a Cobb-Douglas functional form of the

utility function. Their procedure was capable of calculating benefits

 

 

143

directly from the utility function by deriving information about the
demand for the environmental amenity from the rent function (Freeman,

p. 118, 1979).

Use of Land Rent or Property Value Differentials
The distinction between using changes in land values and land value
differentials as a benefit measure is an important one. Freeman pointed
this out in his comment on the Lind article. He stated that an upper
bound estimate of the benefits from land enhancement can be found by
comparing the value of an unimproved site with the value of a similar
site that has already been improved. Freeman went further to say that
property enhancement benefits could, therefore, be determined from pro-
perty rent differentials derived from cross-section analysis. He also
noted that in contrast to the change in property value approach the re-
sult mentioned above for differentials does not depend on fulfilling the
zero surplus assumption and does account for relocation effects (Freeman,
p. 473, 1975).
Freeman (pp. 113-114, 1979) restated the results when property value
differentials are used with the following equation.
B = s(r2 — r1) Equation (7)
Where:
B = benefits
5 = number of acres polluted of the total acreage (n)
r2 = rents on the polluted land
r1 = rents on the unpolluted land
The above equation says that benefits can be determined by comparing the

unpolluted and polluted land rents before the environmental improvement.

 

144

This result is consistent with the earlier results of Lind except that
the equality in the equation is incorrect. The equality should be al-
tered as follows in order to be consistent with results of Lind and
earlier results of Freeman.

B s s(r2 - r1) Equation (8)
Equation (8) states that the rent differentials existing before an envi-
ronmental improvement provide an upper bound estimate on the benefits
of the project that caused the improvement. Once this correction is
made the statement that the results from using such differentials or
rent changes do not depend on the zero surplus assumption and need not
take into account relocation effects can be made.

The reason the distinction between observing property value changes
and observing property value or rent differentials was mentioned was to
emphasize that different assumptions are required for each. The use of
property rent differentials has been more common in practice since it
offers an estimate of what the value of an environmental improvement
would be beforehand (ex ante). The change in property value approach
can also be used in an ex ante sense to measure the benefits of a pro-
ject with information from a similar project. However, it is not appli-
cable if the proposed project or other factors in the region affecting
price are different. The use of differentials usually has the advantage
of a larger sample size which improves the credibility of results. The
idea of observing property rent differentials has been applied by many
researchers in studies attempting to estimate the benefits of air quality
improvements. The next section elaborates on these studies.

The second reason the distinction between observing property rents

differentials and changes in property rents or values is that the results

145

from observing rent differentials are consistent with the hedonic price
technique (Freeman, p. 114, 1979). The use of rent differentials can

be considered a simple form of the hedonic approach (Freeman, 1980).

Property Value Differentials and Air Pollution: Evolution of the Hedonic

 

Price Technigue

Ridker was the first economist to use land value information in an
attempt to determine the value of changes in the environment. His 1967
research utilized residential property value data to measure the benefits
of air pollution abatement. Ridker and Henning (p. 253, 1967) used
cross—sectional residential property value data from census tracts in
St. Louis in a study of air pollution effects on land values. The re-
sults indicated that air pollution was a significant variable in explain-
.ing property value. However, the conclusions that they made about their
regression coefficients aroused considerable interest and subsequent
debate. Specifically, the authors noted that the coefficient of a pol-
lutant measure (sulfation level) was significant. They then concluded
that if sulfation levels were reduced by .25 milligrams/100 centimetersz/
day over the entire metropolitan area, then property values could increase
by up to 82,790,000 dollars. This figure was calculated using the regres-
sion coefficient for sulfation level. Ridker and Henning stated that
households should be willing to pay at least this amount for the speci-
fied pollution reduction.

Freeman (p. 415, 1971) commented on Ridker and Henning's study. He
stated that they had over interpreted their regression results. A regres—
sion equation can only explain variation in mean property values among

observations. The coefficient of an air pollution measurement variable

 

 

146

can predict the differences between two property's value, ceteris
paribus. In other words, everything else, including air quality over
the rest of the metropolitan area, must be held constant. This means
that the regression equation cannot predict resulting patterns of pro—
perty values (or even the change in value of one land parcel) when air
quality over the entire metropolitan area has changed. The reason can
be understood by considering the supply and demand forces involved in
the relationship between property values and air quality in an urban
area. This relationship involves interaction between the availability
of land with various air quality levels (supply) and tastes, preferences,
prices of other goods, and income (demand). For a given set of factors
that influence demand, the different factors influencing supply of land
with various air quality levels will cause different property value pat—
terns. This can also be said for demand changes with supply factors
held fixed. Thus, the regression results would differ from the true
property value changes if any demand or supply factors changed (Freeman,
p. 415, 1971). This is precisely the error made in the Ridker and
Henning study. They used their regression results to predict the pattern
of property values that would exist over the entire city if the supply
of clean air was increased. The error was ignoring the fact that demand
and supply interact to yield the resultant pattern of property values in
the city.

Freeman (p. 416, 1971) called for the development of a general
equilibrium model that could be solved for the pattern of property values
or rents as a function of air quality patterns and other important fac-

tors. He stated that cross—section regression coefficients for air

 

147

quality variables are incapable.of identifying the demand curve for clean
air. His reasoning is that such coefficients reflect both supply and
demand elements.

Anderson and Crocker (pp. 471—472, 1971) studied the relationship
between air pollution and residential property values using cross—
sectional data from three U.S. cities. One objective of their research
was to examine the theoretical background of cross—section demand studies.
The authors' proposed model was derived from the "new” theory of consumer
behavior of Lancaster (1966). This theory suggests that an individual's
utility is derived from the characteristics embodied in goods rather than
the actual goods themselves. Thus, the willingness to pay for a parti-
cular good depends on the particular goods characteristics and the income
of the purchaser (Anderson and Crocker, p. 471, 1972). The authors suggest
that this demand model provides a theoretical rationale for using regres-
sion procedures that were used by Griliches (1961) to determine hedonic
price indexes for automobiles (Anderson and Crocker, p. 172, 1971).

Anderson and Crocker interpreted their regression equations as bids
for properties dependent on property characteristics and income. There-
fore, derivatives with respect to any characteristics represent a change
in bid. The authors disagree with Freeman's contention that there is an
identification problem in air pollution/property value studies resulting
from supply and demand forces. The authors state that a general equili—
brium model does exist that can be used in principle to produce the new
pattern of property values after an environmental change from cross-
sectional analysis. However, they note that calculations using this
model are difficult. Lastly, Anderson and Crocker argue that air

pollution/property value studies are useful even if specification of a

148

general equilibrium model is not possible. The reasoning offered is

that such studies can provide at least a partial estimate of willingness

to pay to avoid environmental degradation. The authors note the statis-

tical significance of the relationship between air pollution and property
values found by most researchers. Finally, Anderson and Crocker surmise

that studies of this type are the only "hard" evidence obtained on will-

ingness to pay to avoid pollution.

Small (p. 105, 1973) added some clarity to the interpretation of
results from air pollution/property value studies. He concurred with
Freeman that regression results from such studies cannot isolate demand
from supply elements. However, he believes that much information can be
gained from regression results because they represent equilibrium situ—
ations. Specifically, he states that the pollution coefficient in a
regression equation reflects the marginal valuation by consumers on uni-
form pollution level changes.

The assumptions needed for the above result are few. One assumption
needed is the existence of many housing types that offer a continuum of
pollution levels. A second assumption is that there are many consumers
in short run equilibrium with a fixed housing supply. Each consumer
maximizes his utility (u) which depends on his income net of rent paid
for housing (m), pollution level at residence (p), and a vector of
characteristics of a property (h), subject to his income level (y).

Also, a government pollution abatement paramedn‘A exists. Higher values
of A imply lower uniform pollution levels.

If the above assumptions are met, equilibrium rents are determined
by equating supply and demand at each location. Since consumer's utility

is dependent only on p and h, the rent schedule for the area can be

 

 

149

described by the equilibrium rent function r(p,h,A). The function can
be estimated empirically. Now, consider the utility maximization deci—

sion of a consumer.

max u(m,p,h) Equation (9)
Subject to
m + r (p,h,A) = y Equation (10)

The first order conditions for the maximization problem are:

 

bp
:zjgm = f: Equation(11)
2373; ='%% Equation(12)

The left hand side of Equation (11) can be interpreted as the marginal
benefit of pollution abatement by each consumer residing at a location
described by the particular p and h. The right hand side equality is
the slope of the regression equation. The important conclusion of this
analysis is that if the slopes are aggregated over all consumers that
the result is the total marginal benefit of pollution abatement (Small,
p. 106, 1973).

Small points out that the calculation above to estimate aggregate
marginal benefits is exactly the calculation made by Ridker and Henning
in their study. However, he notes that Ridker and Henning interpreted
their result as a prediction of aggregate property value changes.
Instead, the calculation should have been interpreted as a welfare mea-
sure (Small, p. 106, 1973).

Small also points to the unrealistic assumptions made by Anderson
and Crocker that all individuals in a given income class have identical

utility functions. He suggests that in an urban area we are observing

 

 

150

a single housing market. The differences in choices are caused by
different preferences rather than income constraints.

Another shortcoming that Small notes in the Anderson and Crocker
study that income is used as an explanatory variable. He argues that
since income is determined endogenously by the individual household the
pollution coefficient will be biased toward zero.

Lastly, Small comments on the use of the regression equation to
predict the results from large changes in the environment. He states
that the single regression equation cannot be interpreted as a demand
schedule because there are no empirical data available to give us infor-
mation on individuals' demand schedules outside the range of actual
experience. However, we can measure some characteristics of consumers'
preferences. Thus, we are able to use this information to value marginal
changes in environmental quality. The important point is that a regres-
sion equation cannot be used to predict the response of consumer or group
of consumers to large environmental changes. The regression equation
does predict the response to small environmental changes by individuals
having a large range of income and preferences (Small, p. 106, 1973).

Freeman (1974), in an additional comment on air pollution/property
value studies, proposed a model from which he drew conclusions similar
to Small. Freeman's analysis demonstrates that the estimated partial
relationship between air quality and property values cannot be inter-
preted as the demand curve for clean air. Instead, he contends that the
relationship in this rent function is an opportunity locus or quasi-
supply curve for households. From this information it is possible to
estimate the aggregate marginal willingness to pay for an air quality

improvement. The manner in which this benefit measure can be detennined

 

 

151

is by analyzing the slope of the partial rent function. The slope of
this function is the marginal purchase price function which is a locus
of the equilibrium marginal willingness of all households to pay. These
marginal willingness to pay can be summed to find the aggregate marginal
willingness to pay or marginal benefit function. However, the benefits
from non—marginal air quality changes are not obtainable unless more
information is provided about the shape of the marginal willingness to
pay curve (Freeman, p. 555, 1974A). A more detailed discussion of

Freeman's reasoning is in the next section.

Hedonic Price Technique

 

The hedonic price technique is a method capable of estimating the
implicit prices of characteristics of goods. The goods in question must
be slightly differentiated but still in the same product class. The
implicit prices of characteristics are called hedonic prices. Hedonic
housing prices have been used to estimate the benefits and demand for
land and environmental improvements. The procedure used for the estima—
tion of benefits like air pollution abatement involves two distinct
steps. First, the implicit price of the characteristic provided by the
project is estimated from cross-sectional data. This step produces
estimates of the benefits or marginal changes in the level of a parti-
cular characteristic. Secondly, the demand curve for the characteristic
is found by regressing the implicit prices of characteristic against
observed quantities of the characteristic. The second stage can lead
to estimation of the benefits of non-marginal changes in levels of the

characteristics (Freeman, p. 78, 1979).

 

 

152

Theory of the Technique. The central hypothesis of the hedonic

 

price technique is that commodities are valued by individuals for the
utility bearing characteristics they possess (Rosen, p. 34, 1974). As
mentioned above, the prices of the characteristics are termed hedonic
prices.

The two economists that introduced ideas that are important to the
hedonic method were Griliches and Lancaster. On the empirical side,
Griliches (1961) utilized a multiple regression hedonic approach to the
construction of automobile price indexes that accounted for qUality
change. His approach to index construction was based on the rationale
that many varieties of a particular commodity can be comprehended in a
smaller number of basic characteristics. Thus, the commodity can be
viewed as an aggregate of individual components or characteristics
(Griliches, p- 4, 1971).

Lancaster (1966) formulated a new approach to consumer theory in
which he suggested that it was the properties or characteristics of goods
that individuals derive utility from rather than the goods themselves.
Thus, goods can be thought of as bundles of characteristics.

The first instance where the ideas of Griliches and Lancaster were
explicitly used in estimating benefits of environmental change was in
Anderson and Crocker's (1971) research. As noted earlier, this effort
investigated the relationship between air pollution and property values
as observed from cross-section data.

Freeman has provided many insights into the use of the heconic tech-
nique for the valuation of environmental improvements. Specifically,

Freeman has shown how the theory of the hedonic method can be applied to

 

 

 

153

the residential housing market and developed procedures for benefit
estimation. A discussion of Freeman's procedure is presented below.
Housing is defined as a product class that is differentiated by
characteristics such as number of rooms, size of lot, location, water-
frontage, and numerous other attributes. The price of a particular home
can be considered a function of its structural (S), neighborhood (N),
and environmental (E) characteristics. Thus, any unit of housing (Ht)

can be thoroughly described by a vector of its characteristics.

PHt H (S.. ., N. ..., E.. ..) Equation (13)

The equation above can be called the hedonic or implicit price function.
The first step in the hedonic technique involves the estimation of this
function from observations of prices and characteristics of different
housing units. Once the function is estimated, the price of any parti-
cular unit can be calculated if the characteristics are known.

Step 1 of the technique also includes the derivation of the marginal
implicit price function. This is performed by taking the partial deri-
vative of the above estimated function with respect to a particular
characteristic. Suppose this characteristic is air quality (EA). Equa-
tion (14) illustrates the marginal implicit price function for air

quality.
aPHt/aEA = PH (EA) Equation (14)

The marginal implicit price function gives the increase in expenditure
on housing that is required to obtain a unit more unit of air quality,
ceteris paribus (Freeman, p. 156, 1979A).

Suppose that the partial (all else constant) relation between EA

and PH (Equation 13) has been estimated for an urban area. The relation

 

154

for a nonlinear case is represented by Figure B-1. Figure B-2 represents

the marginal implicit price relationship of Equation 14. A needed

assumption is that households are price takers in the housing market.

As such, the household can be visualized as facing an array of implicit

marginal price schedules. The maximization of household utility can be

viewed in the following way. Each household simultaneously moves along

each marGinal implicit price schedule until it reaches a point where

its marginal willingness to pay for an additional unit of a character-

istic just equals the marginal implicit price of that characteristic. '
In equilibrium, the marginal implicit prices of attributes of a housing ‘
unit must be equal to the marginal willingness to pay for those attri-

butes. Figure B-2 illustrates the marginal implicit price of EA (PE (EA)L

Also, the marginal willingness to pay curves for two households as well

as the equilibrium choices of Ey and EZ of the two households are shown.

Households (such as y or 2) choose a site where its marginal willingness

to pay for EA is equal to the marginal implicit price of EA (Freeman,

p. 157, 1979A).

The first stage of the hedonic technique does not directly identify
the inverse demand schedule for EA‘ Under certain conditions, a second
step can utilize information from the hedonic price function to estimate
the demand relationship for a characteristic. Several assumptions are
required. First, assume each individual purchases only one unit of
housing. Also assume that individuals utility functions are separable
in housing units and its characteristics (Freeman, p. 919, 19798). The
second step combines implicit price and quantity information with other

factors influencing demand for EA such as the level of EA’ income (y),

 

 

155

Figure B-l
The Hedonic Price Function

 

 

Level of Environmental Characteristic

Figure B-2
The Marginal Implicit Price Function and Willingness to Pay Curves

  

P ( E )

_ . 5 A

u (L )
Z n

H

 

Ey E Level of Environnental Characteristic

 

Source for Figures: Freeman, p. 124, 1979.

 

156

and other socioeconomic characteristics of households. Equation (15)
describes this willingness to pay relationship.
Wi = w(EA1, yi, ...) Equation (15)

The observed PE (EA) of each household is used as a measure of wi.
A

The estimation of Equation (15) is a necessary step for estimating
the demand curve for EA. Two special cases are worth noting. First, if
the hedonic price function is linear in EA’ then the demand curve cannot
be found since the marginal implicit price is constant. Secondly, if all
households have identical utility functions and incomes, then Equation 14
(the marginal implicit price function) is the inverse demand curve.
Equation 14 is a locus of points on households marginal willingness to
pay curve. Since all individuals are identical, all these points lie on
the same marginal willingness to pay curve (Freeman, p. 158, 1979A).

If neither of the cases above applies, the examination of the supply
side of the market for the characteristic is required. Three situations
are relevant. The first case is where the supply of houses with given
characteristics is perfectly elastic. Here, the implicit price function
is considered as exogenous to households. A regression of observed levels
of the attribute against observed implicit price as defined in Equation
14, income, and other characteristics of the household should reveal the
inverse demand curve. This case is not typical of housing. The second
situation is when the supply of housing units is fixed. In this case,
households can be viewed as bidding for fixed quantities of models with
certain bundles of attributes. The inverse demand curve can be obtained
by regressing each household's marginal willingness to pay (measured by
its implicit price) against the level of the characteristic chosen,

income, and other variables. The third situation occurs when both

157

quantities supplied and demanded of attributes are functions of prices,
a simultaneous equation approach much be used (Freeman, p. 158, 1979A).
Nelson (1978) conducted one of the few air quality studies adopting the
simultaneous equation approach.

Benefit measurement once the inverse demand curve is identified is
straightforward. The benefits from changes in the environment are cal-
culated by measuring the areas under the demand curve between the old
and new quantity of the environmental characteristic. The demand curves
are not income-compensated so further consideration must be given to the

various measures of consumer surplus (Freeman, p. 920, 19798).

 

APPENDIX C
DATA MODIFICATION INFORMATION “I

 

 

158

Appendix C

Data Modification Information—-Use of Cost

Trend Multiplierslto Adjust Value of Improvements
(in 1972 Costs) for Time

Table C—l
Multiplier Adjustment Calculations

 

 

 

Year Multiplier Factor (1972) - Factor (year) Factor
1981 1.000 2.000 - 0.000 = 2.000
1980 1.065 2.000 — 0.065 = 1.935
1979 1.161 2.000 - 0.161 = 1.839
1978 1.265 2.000 - 0.265 = 1.735
1977 1.391 2.000 - 0.391 = 1.609
1976 1.489 2.000 — 0.489 = 1.511
1975 1.581 2.000 - 0.581 = 1.419
1974 1.698 2.000 - 0.698 = 1.302
1973 1.838 2.000 - 0.838 = 1.162
1972 2.000 2.000 — 1.000 = 1.000
1971 2.208 2.000 - 1.208 = 0.792
1970 2.350 2.000 - 1.350 = 0.650

Example calculation - Value of improvements = $10,000 (1972)
Value in 1978 = 10,000*1.735 = $17,350

 

 

1The cost multipliers are based on a national average from Marshall
and Swift Building Cost Indexes.

Source: Residential Cost Handbook, Marshall and Swift, 1981, p. F-7.

 

 

 

APPENDIX 0
ADDITIONAL INFORMATION ABOUT THE DATA J.

 

      

      

 
   

 

 

TUNA i
,Lfiw

           

 

o_:>wucmcn U
. Q

.IIIIIIIIIIIJIII'J.

O—
c). \ _
Q ,

 

 

 

  

 

 

159
Appendix D

    
    

 

 

 

 

 

 

 

 

 

 

 

 

 

a BEL mm

«fowomx \(Mamq/

int.- MWw:
0

__ we.
1.1.01.-
...31 Va .

Ox _

2:903

93.5%me
h

.

fl

kuoo‘sxmo

. TV

.403

q

 

 

 

cofifn. .

2:92.. o
0x ,4 23m

 

 

 

zanzom

c:m>_oRtb

N i,
f
23:: .

can Eu

m
8

at
own/t ‘

.334 w: 3 V

v

1m. om.

 

rwcwzocm :mmvzuwz :
Hie mgzmwu

_ 1
30m to am

 

 

 

 

 

   

   

 

 

160

Appendix D
Figure 0-2
Clay Township S Area

v

\

 

   

4... ¢.'

54 y n .5 w mnuwza

» \ N L:
1,1,3” Musrgucnr
..
I“ (I A 7

Figure D-3
Putnam Township Study Area

 

 

   
 

. .,’ I . i" -
I ( ; ,,,,, I . ‘5 ‘. .‘ : It'A \ (V ...- r
I . I r-ir . k I ‘
M A . R -- l 0 i 2 "
. I, , . -'
I I >
| I I
i . ..
I .
0 I. . ..... ’1. . .
I ‘ ‘
, ( l
I v r M J‘
I ' u
. .-‘-',1 Jl"i"-- J" a ‘ ' ‘
, . “a“ i I._ . :- ft InICkF/UION ~ I .2
. I V‘ ;. b '10-! )1 “A, '..L—
....-. , ..ar. ...... - J - ‘ 7——. f” " ' 7-7' "' -
... , e “T n- 1.- ri-T‘fil r: 'ItlE‘.Dfi.-& .~ 101
I . . . InirH‘f“ \— nW ‘-

 

,.
O
1-
.‘x
.g,
4
a

I r .

1 A
.\(‘. .
nnolp r.
:-‘ \

a :77”

LAI
-1“:
“v

    

 

 

 

Source: Michigan Department of Conservation (1973).

161

Appendix D

Figure D-4
Harrison Township Study Areas

 

 
   
    
 

 

  
   
       
   
 
  

 

} 3 ‘ .. . ' is“. /
I I ‘ ‘ ‘3'. ‘
I .. ‘ _ I,
C ‘ om: . i . t .
..iy..-rl\_~..-----..u..‘ ‘.nv I -- - - .. vvvvvv
, . u w c .
z. . I . .-. '-l' .. .. '.\ . a . \ / Myth
.. . ..
- ' I | If n - , ~ . .
_ ‘.ifl'r' ’h -$-' .- ..‘r‘rzmq. om IHZ‘M‘J '- ' - .' ' - I ... ._.—..- ...——
l ‘ 'I I ‘
:3 I . .‘ J l . , ‘I .
I I g: I I. . . I... . . .‘ ‘ .”' ....
J. k n. 'L L ‘. q ‘
- f o: ,o-v'~vrbr~0» vor -, I." ...... lav r-o-..
:IACOIBL 3‘ . }. L. . \ “I '
‘ l l- v".-
Y .; . : - ”I . 7"}, / E“). Ennfiytunii
D .‘ _ 7 '.
-~JI----..:..,.....- v A... , .1... ....... .......:.Z. x\
‘ h i
; M n‘. c o M ‘e‘ . ‘
, .
x' ‘I : t‘l'LI'LFtIEL. 0
, 1..:‘_-----‘....-......'..;..-...325. ‘ __,,~,‘
u fi- 9
.' . ~ '1 -- ,
(
z I' “I! .. i ( , CIIESl‘Elu
..-.q..--,...-.;n——....,. Ir...” . .
L I ‘r 71 L‘ j 4
u i- I .- l ‘vn I: v I, .1\, -|'
> 7 I ‘ . L I”, i
. ; I
"""'.'".""""'r"“”:' :--. .tn. ;\ ......” .:'..l-
' . L .. \ :.
i l t i . " O
t ' 'fj , ,/ ‘_. II
n . , ..I

 

 

‘ 4‘ v
'(irlinnl \
mu: Pm" 90v"
., ,q , u -.
WIQLGI ( s...

937'“): ~' EAST
' '/

~.. .1
\' '." . ‘\\\ . . .
TR Pm‘f‘ku‘l/"ll . , .
), U] I».
E Clix?)

I ..n .
Ill mm. ...w \
Ink .4 ).

 

 

SOUTH

 

 

l--II A

DEPARTMENT OT- CON"

/ (it iiié'é‘ts MACOMB c0

 

 

 

 

 

... .... _
.» r;
‘. . ..... .v A
~ i
.W, .
“I ‘ 'J'OCS.‘ Dt-w’t s-onzs
I . g

 

 

 

Source: Michigan Department of Conservation (1973).

162

Appendix D

Table D-l

Names and Units of Variables

 

 

Variable Name

Definition

Measurement Unit

 

 

 

YEAR Year of sale Years (70-80)
VALUI Value of improvements in Dollars in year of
year of sale sale
HZOWFT Waterfrontage on natural or Feet
man—made waterbody
CLNFT Naterfrontage on man—made Feet
channel
NBNFT Haterfrontage on natural Feet
water body
LOTSIZE Area of lot Feet2
DISTNB Distance to natural water Feet
body
DIST Distance to specified refer- Feet
ence point (inland adjacent
lots)
Note: LN if front of variable name indicates the variable has been

transformed by taking the natural logarithm of the value of
the variable.

   

 

 

163
Appendix D
Table D—Z

 

«NVHN. A©HN.q-V AH¢¢.NV ANN“.-V Akmo.fiv Aooo.mv Aoka.m-v Aa=_a>-pv
came“. me Q¢N.MH- w©N¢.HHH Namoooooo. wafi.fi mm.m©wN NL.¢OmHNm. catasuuo-aucea
mammm. Afimm.m-v ANHc.mV Ama.m-v ANHM.NV A©NO.NV Aw=_a>-pv
Hmwmm. we mHH¢.N- wooe.mw mem.fi- mm©.©fikq sm.omomqm p:mom>-matea
~m\mm z mzpmHo Hazomz NH24<> H=s<> x<m> “amatach m_naw.m>

 

and Occupied Land Parcels

“concmamo
”mo mpcwwowwwmoo

 

 

Clay Township Regression Equations for Vacant

   

164

Appendix D

Table D-3
Clay Township Correlation Matrix

 

MNHmHOA

 

Azuongwpaz ou wocmmeo

 

 

 

 

 

 

mmmov.u mZHmHo

Ammmpcogwgmpmzv
omumm. nVOOm.- Hmzom:
mwmmm. ovwoo.- NNNHN. NH24<>
AmucmEm>OLaEH mo warm>v
mHNmH. wwmno. mNNNH. momma. H34<>
wmoHH.- oHonH. oommo. owmom. «Hwom. Nm<m>
MNHmHOA mzpmHo Hazomz H24<> H34<> m<m>

 

 

N m

 

 

 

 

 

 

 

 

 

 

thzmz

 

.Aommpcogwgopmz chzpmzv
Hmzmz

 

AwmmpcoLwLwHaz chcmcuv
mmfinm.n Hngu

 

Ampcmsm>ogqsm we wsz>v

 

 

 

vmmmH. mmuvo.u H24<>
«mmofi.u NHvoH.u ammm¢. Nm<m>
HmnoH.u mweofi.u omew. Nmmmm. , m<m>
Hmzmz hu34u H=4<> Nm<m> z<m> w

 

 

 

 

 

 

 

 

 

 

166

Appendix D

Table D-5
Harrison Township East (2b) Correlation Matrix

 

Azuongmpmz ow wucmpmva mOJV
mzhmHo 24.

 

Ammmucogwgmpmz mOJV

 

 

 

 

 

mmomfi.- Hmzomz :4
AmpcmEm>ogaeH +0 mzpm>v
mmNNH.- mmomo. H34<>
mmmwo. mwmmm.- Haqu. Nm<m>
Nwmqo.‘ ommmm.- owfimw. Nwmmm. m<m>
mzhmHo web Huzomx web H34<> m<m> m<m>

w

 

 

 

 

167

Appendix D

Table D-6
Harrison Township South Correlation Matrix

 

 

 

 

 

 

 

MNHmHog
N
mkokm. mNHmHOA
Awucmpmwo nebv
mkmvv.- mowom.- HmHQZA
Awmmucogwgmpmzv
mwmmm.- mmomm.- Hmoom. Hezomx
allllllllllllllllllllllll
AmpcwEo>ongH we m:_m>v
mmvmfi. wwmom. om0mm.- Nm¢¢N.- H24<>
ovomfi.- qumo.- ammfi. ammoo. mmomn. mm<m>
wmeH.- wuooH.- cmmfio.- mNoH. mmmmu. wwmmm. m<m>
NMNHWHOA mNHmHOA HmHQGOA Fuzomz H24<> mm<m> m<m>
_ L

 

 

 

 

 

 

 

 

.

.../.3.

168

Appendix D

Table D-7
Livingston Township Correlation Matrix

 

 

 

 

 

 

 

 

 

 

 

‘1
mNHmHOAZA
Illlllnllllllllllllllulll
. Ammmpcoxwgwpmz webv
moemo - Hazomzza
oofimm. moHHH. NH24<>
AmpcwEm>oLQEH +0 m:_m>v
omewfi. HHmNH. vmmmm. H24<>
Hmkmo. mmHHH. owmfim. Nowmm. Nm<m>
Hemoo. meHH. mqwfim. onmm. mmmmm. m<m>
MNHmHOA we; Hmzomz GOA NH=4<> H34<> Nm<m> m<m>

 

 

 

   

 

169

Appendix D

Table D-8
Description of Joint Test of Significance

Harrison Township South Study Area (Equation 4)

 

General Linear Statistical Test1

HO : Bq = Bq+1 = ...... F = 0

HA : Not all the B's equal zero

Where: The B's are coefficients of variables in the full model
but not in the restricted model.

F* = SSE {Restricted Model) - SSE (Full Model) I SSE {Full Model)
n-q). - (n-p) ' n-p

 

 

 

or
* = .-
F SSR (X ...... , Xp_1/ X1 ...... , Xq_1) . MSU (Full Model)
(q-p)
Nhere:
n = number of variables in full model
q = number of variables in the restricted model
SSR = Sum of squares due to regression
SSE = Sum of squares due to error
1

See Neter and Wasserman, p. 264, 1974.

   

170

Appendix D

Table D—9
Joint Test of Significance--Test 1

 

 

TEST 1
Restricted Model: Price = FO + BlYEAR + BZYEAR2 + B3VALUI
+ B4LOTSIZE + BSLOTSIZEZ
Full Model: Price = BO + BIYEAR + BZYEAR2 + B3 VALUI + B4L0TSIZE
+ BSLOTSIZE2 + B6LOGDISTANCE + B7H20NFT
H0 B6 = B7 = 0
HA: Not both B6 and B7 equal zero
Test:
* _ .
F — SSE(Restr1cted) — SSE(Full) MSU(Full)
(p-q)
* -
F — 16179826859 é 10533670697 e 123925538
* _
F ‘ Eggs—6153 123925538

*

F = 22.8
Compare to F(1 -«, p - q, n - p)
if d.= .01, then F(.99, 2, 85)
F(.99, 2, 120) = 4.79

II
C

Since F* is greater than 4.79, Reject HO that B6 = B7

171

Appendix D
Table D-10
Joint Test of Significance—-Test 2
TEST 2
Restricted Model: Price = F0 + BIYEAR + BZYEARZ + B3VALUI
+ B4LOGDISTANCE + BSHZOWFT

Full Model: Price = B0 + BlYEAR + BZYEAR2 + B3VALUI + B4L0GDISTANCE

+ BSHZONFT + B6L0TSIZE + B7L0TSIZE2
H B = B = 0

HA : Not both B6 and B7 equal zero

 

 

Test:
F* = SEEfRestri%:ed)q3 SSE(Full) é MSE(Full)
F* = 11023893223 5 10533670697 123925538
F* = £292§Z§§Z 123925538
F* = 245111263 e 123925538
F* = 1.9778

Compare to F(1 - , D - q, n - P)
if = .10, then F(.90, 2, 85)
F(.90, 2, 120) = 2.35
0 that B6 = B7 = 0

However, LOTSIZE and LOTSIZE2 were included in the regression
equation due to their logical signs and their improvement of
R2 and R2.

Since F* is less than 2.35, Accept H

172

Appendix D

Table D-11
Development Cost Calculations-—Clay and Harrison Township East

Cost Assumptions: based on brief survey of contractors
Seawall - $75.00/foot Seawall (steel)
Dredging cost - approximately $1.00 cubic yard
Dimensions - depth of channel - 15 feet including 2 feet above water

Additional fill - not costed

Clay Township Calculations

 

Typical lot dimensions = 75 feet * 120 feet
Attributing half of channel width (40 feet) to each lot
Dredging cost = length * width * depth
13.33 yds * 25 yds * 5 yds = 1666.67 cubic yds
1666.67 cubic yds $ $1.00 cubic yds = $1666.67
Seawall cost = 75 feet * $75.00/foot = $5625.00
Total $7291.67

(rounded to $7500.00)
Harrison Township East Calculations
Typical lot dimensions = 50 feet * 120 feet
Attributing half of channel width (50 feet) to each lot
Dredging cost = 16.67 yds * 16.67 yds * 5 yds = $1388.88
Seawall cost = 50 feet * $75.00/foot = $3750.00
Total $5138.00

(rounded to $5500.00)

 

173

Appendix D
Table D-12
Development Cost Calculations-—
Harrison Township South and Putnam Township
Harrison Township South
Typical lot dimensions = 100 feet * 120 feet
Attributing half of cannel width (40 feet) to each lot
Dredging cost = length * width * depth
13.33 yds * 33.33 yds * 5 yds = 2222 cubic yds
2222 cubic yds * $1.00/cubic yd = $2222.00
Seawall cost = 100 feet * $75.00/foot = $7500.00
Total $9722.00

(rounded to $9500.00)

Putnam Township
Typical lot dimensions = 60 feet * 150 feet
Attributing half of channel width (40 feet) to each lot
Dredging cost = length * width * depth
13.33 yds * 20 yds * 5 yds = 1333.33 cubic yds
1333.33 cubic yds * $1.00/cubic yds = $1333.33
Seawall cost = 60 feet * $75.00/f00t = $4500.00

Total $5833.33

(rounded to $6000.00)

BIBLIOGRAPHY

 

 

BIBLIOGRAPHY

 

Anderson, Lee G. and Russell F. Settle, Benefit-Cost Analysis: A Practical
Guide. Lexington Books, Lexington, MA, 1977.

Anderson, Robert J. and Thomas D. Crocker, "Air Pollution and Property
Values: A Reply.” Review of Economics and Statistics, Vol. 54,
No. 4, pp. 470-473, November, 1972.

 

"Air Pollution and Residen—
tial Property Values." Urban Studies, Vol. 8, pp. 171-180,
October, 1971.

 

Bartlett, Randall, Economic Foundations of Political Power. The Free
Press, New York, NY, 1973.

 

Bhattacharyya, Gouri K. and Richard A. Johnson, Statistical Concepts and
Methods. John Wiley and Sons, New York, NY, 1977.

 

Bishop, Richard C. ”Endangered Species, Irreversibility and Uncertainty:
A Reply." American Journal of Agricultural Economics, pp. 376-379,
May, 1979.

 

Boadway, Robin W. Public Sector Economics. Winthrop Publishers, Cam—
bridge, MA, 1979.

 

Boto, K. G. and William H. Patrick, Jr. ”Role of Wetlands in the Removal
of Suspended Sediments," Wetland Functions and Values: The State of
Our Understanding. P. E. Greeson, J. R. Clark, and J. E. Clark,
eds., American Water Resources Association, Minneapolis, MN, pp. 479-
489, 1979.

 

Brande, Justin, ”Worthless, Valuable or What? An Appraisal of Wetlands,‘I
Journal of Soil and Water Conservation, pp. 12-16, January-February,
1980.

 

Burton, Thomas M. The Effects of Riverine Marshes on Water Quality,
Paper presented at Midwest Conference on Wetland Management and
Values, St. Paul, MN, June 17-19, 1981.

 

Department of the Army, Corps of Engineers, ”Regulatory Program of the
Corps of Engineers.” Federal Registe , pp. 37121-37164, July 19,
1977.

174

 

175

Department of the Army, Corps of Engineers, ”Permits for Activities in
Navigable or Ocean Waters." Federal Register, pp. 31320-31344,
July 25, 1975.

Dwyer, John, John R. Kelly, and Michael D. Bowes, Improved Procedures
for Valuation of the Contribution of Recreation to National Econ-
omic Development, University of Illinois, Water Resources Center,
1977.

 

Executive Order 11990: Protection of Wetlands, Federal Register, Vol. 42,
No. 101, pp. 26961—26965, May 25, 1977.

Fetter, C. W., W. E. Sloey, and F. L. Spangler, "Use of a Natural Marsh
for Wastewater Polishing." Journal of the Water Pollution Control
Federation, Vol. 50, pp. 290-307, 1978.

Foster, John H. "Measuring the Social Value of Wetland Benefits,"
Wetland Functions and Values: The State of Our Understanding. P. E.
Greeson, J. R. Clark and J. E. Clark, eds., American Water Resources
Association, Minneapolis, MN, pp. 84-92, 1979.

 

Freeman, A. Myrick, Personal Communication, August 11, 1980.

"Hedonic Prices, Property Values, and Measurement
of Environmental Benefits: A Survey of the Issues.” Scandinavian
Journal of Economics, pp. 154-173, 1979A.

 

”Approaches to Measuring Public Goods Demands."
American Journal of Agricultural Economics, Vol. 61, pp. 915-920,
December, 1979B.

 

. The Benefits of Environmental Improvement, Resources
for the Future, John Hopkins University Press, Baltimore, MD, 1979.

 

. ”Spatial Equilibrium, the Theory of Rents, and the
Measurement of Benefits from Public Programs: A Comment.” Quarterly
Journal of Economics, Vol 89, No. 3, pp. 470—473, 1975.

 

. ”On Estimating Rfllution Control Benefits from Land
Value Studies.” Journal of Environmental Economics and Management,
Vol. 1, pp. 74-83, 1974.

 

”Air Pollution and Property Values: A Further Com-
ment.“ Review of Economics and Statistics, Vol. 53, No. 4, pp. 554-
556, 1974A.

 

“Air Rfllution and Property Values: A Methodological
Comment." Review of Economics and Statistics, Vol. 53, pp. 415-416,
November, 1971.

 

Goemaere-Anderson Wetland Protection Act of 1979, Michigan Public Act
203 of 1979, pp. 1-6.

 

176

Goldstein, Jon H. Competition for Wetlands in the Midwest, Resources
for the Future, Washington, D.C., 1971.

 

Gordon, H. Scott, ”The Economic Theory of a Common Property Resource:
The Fishery.” Journal of Political Econpmy, Vol. 62, No. 2,
pp. 124-142, 1954.

 

Gosselink, James 0., Eugene P. Odum and R. M. Pope, The Value of the
Tidal Marsh, Publication N., LSU-56-74-03, Center for Wetland Re-
sources, Louisianna State University, Baton Rouge, LA, 1974.

Griliches, Zvi, Price Indexes and Quality Change. Harvard University
Press, Cambridge, MA, 1971.

Hedonic Price Indexes for Automobiles: An Econometric

Analysis of Quality Change, 1961.

Hardin, Garrett, ”The Tragedy of the Commons.“ Science, Vol. 162,
pp. 1243-1248, 1968.

Haveman, Robert H. The Economics of the Public Sector, Introduction to
Economics Series, Kenyon A. Knopf, ed., John Wiley and Sons, Santa
Barbara, CA, 1976.

 

Howe, Charles W. Natural Resource Economics: Issues, Analysis, and
Policy. John Wiley and Sons, New York, NY, 1979.

 

Jaworski, Eugene and C. Nicholas Raphael, Fish, Wildlife, and Recrea-
tional Values of Michigan's Coastal Wetlands, U.S. Fish and Wild—
life Service, Twin Cities, MN, 1978.

 

 

Kadlec, Robert H. "Wetlands for Tertiary Treatment.“ Wetland Functions
and Values: The State of Our Understanding, P. E. Greeson, J. R.
Clark and J. E. Clark, eds., American Water Resources Association,
Minneapolis, MN, pp. 490-504, 1979.

 

Kadlec, R. H. and J. A. Kadlec, ”Wetland and Water Quality.” Wetland
Functions and Values: The State of Our Understandipg, P. E. Greeson,
J. R. Clark and J. E. Clark, eds., American Water Resources Associ-
ation, Minneapolis, MN, pp. 436-456, 1979.

 

Lancaster, Kelvin J. ”A New Approach to Consumer Theory.” Journal of
Political Economy, Vol. 74, pp. 132-157.

Libby, Lawrence W. The Role of the University Social Scientist in Devel-
ppment and Implementation of Environmental Policy at the National
Level, Paper presented at Internal Conference on Rural Development
at BHckaskog, Sweden, June 23-30, 1981.

 

 

 

Lind, Robert C. "Spatial Equilibrium, The Theory of Rents, and the
Measurement of Benefits from Public Programs.“ Quarterly Journal
of Economics, Vol. 87, pp. 188-207, May, 1973.

177

Luken, Ralph A. Preservation Versus Development: An Economic Analysis
of San Francisco Bay Wetlands. Praeger Publishers, New York, NY,
1976.

 

 

Mishan, Edward J. Cost-Benefit Analysis. Praeger Publishers, New York,
NY, 1976.

Mitsch, William J., Max D. Hutchison, Gerald A. Paulson, and George
Benda, The Momence Wetlands of the Kankakee River in Illinois,

Illinois Institute of Natural Resources, Chicago, IL, 1979.

Morrison, David, The Impacts of Development of Wetlands on Water Quality,
Report prepared by Southeast Michigan Council of Governments, June,
1979.

 

Moser, C. A. and G. J. Kalton, Survey Methods in Social Investigation.
2nd edition, London, 1972.

 

Murray, D. Kenneth, Public Policies and Programs Affecting the Future Use
of Michigan's Coastal Wetlands, Unpublished technical paper, Depart-
ment of Resource Development, Michigan State University, East
Lansing, MI, 1980.

 

 

 

Musgrave, Richard A. and Peggy A. Musgrave, Public Finance in Theory and
Practice. 2nd edition, McGraw—Hill, New York, NY, 1976.

Natural Resource Defense Fund v. Calloway, 392 F. Supp. 685 (1975).

 

Nelson, Jon P. Lecture delivered in Environmental Economics course at
Pennsylvania State University, University Park, PA, Spring, 1979.

”Residential Choice, Hedonic Prices, and the Demand for
Urban Air Quality" Journal of Urban Economics, Vol. 5, pp. 357-
369,1978.

 

Neter, John and William Wasserman, Applied Linear Statistical Models:
Regression, Analysis of Variance, and Experimental Designs.
Richard D. Irwin, Inc., Homewood, IL, 1974.

 

 

Nicholson, Walter, Intermediate Microeconomics and Its Application. 2nd
, The Dryden Press, Hinsdale, IL, 1979.

 

. Micro Economic Theory: Basic Principles and Exten—
tions. 2nd ed., The Dryden Press, Hinsdale, IL, 1978

 

 

Park, William M. and Sandra S. Batie, Methodolo tical Issues Associated
with Estimation of the Economic Value of Coastal Wetlands in Improv-
ing Water Quality. Sea Grant Research Report No. VPI- SG- 79-09,
Virginia Tech, Blacksburg, VA, 1979.

 

 

Park, William M. and Burl F. Long, "Issues Associated with Wetlands
Management.” Proceedings of Conference on Current Land and Water
Resource ISSUes in Virginia, Virginia Polytechnical Institute and
State University, September, 1979.

 

 

178

Pittman Robertson Act (Wildlife Restoration Act of 1937), Public Law
No. 415, Ch. 889, 50 Stat. 917, September 2, 1937.

Platt, John, "Social Traps.“ American Psychologist, pp. 641-651, August,
1973.

Polinsky, A. Mitchell and Daniel L. Rubinfeld, “Property Values and the
Benefits of Environmental Improvements: Theory and Measurement,"
Lowdon Wingo and Alan Evans, eds., Public Economics and the Quality
of Life, John Hopkins University Press, Baltimore, MD, pp. 154-180,

77.

Polinsky, A. Mitchell and Steven Shavell, ”Amenities and Property Values
in a Model of an Urban Area." Journal of Public Economics, Vol. 5,
No. 1-2, pp. 119-129, 1976.

. "The Air Pollution Property
Value Debate." Review of Economics and Statistics, Vol. 57, No. 1,
pp. 100—104, 1975.

 

Public Law 92-500, Section 404, 33, U.S.C. 1344.

Reid, George K. and Richard D. Wood, Ecology of Inland Waters and Estu-
aries. 2nd edition, D. Van Hostrand Company, New York, NY, 1976.

Ridker, Ronald G. Economic Costs of Air Pollution: Studies in Measure-
ment. Praeger, New York, NY, 1967.

Ridker, Ronald G. and John A. Henning, ”The Determinants of Residential
Property Values with Special Reference to Air Pollution." Review
of Economics and Statistics, Vol. 49, No. 2, pp. 246-257, 1967.

 

Rosen, Sherwin, "Hedonic Prices and Implicity Markets: Product Differ-
entiation in Pure Competition." Journal of Political Econgmy,
Vol. 82, No. 1, pp. 34—55, 1974.

Schmid, A. Allan, Property, Power, and Public Choice: An Inquiry into
Law and Economics. Praeger Publishers, New York, NY, 1978.

”Predicting the Performance of Alternative Institu-
tions.|I Law and Economics: An Institutional Perspective, Warren J.
Samuels and A. Allan Schmid, eds., Martinus Nijhoff Publishing,
Boston, MA, 19808.

The Political Economy of Public Investment, unpub-
lished manuscript for public project analysis course taught at
Michigan State University, East Lansing, MI, Spring 1980A.

Schmid, A. Allan and James D. Shaffer, "Community Economics: A Framework
for the Analysis of Community Economic Problems,” unpublished paper
prepared for Public Affairs Management 201, Michigan State Univer-
sity, East Lansing, MI, 1979 edition.

 

 

 

179

Seneca, Joseph J. and Michael K. Taussig, Environmental Economics, 2nd
edition, Prentice Hall, Englewood Cliffs, NJ, 1979.

Shabman, Leonard A. and Sandra S. Batie, Basic Economic Concepts Import-
ant for Wetlands Valuation, paper presented at Midwest Conference
on Wetlands Management and Values, St. Paul, MN, June 17—19, 1981.

 

Estimating the Economic Value
of Coastal Wetlands: Conceptual Issues and Research Needs. Sea
Grant Project Paper, VPI-SG-79—08, Department of Agricultural Econ-
omics, Virginia Tech, Balcksburg, VA, 1979.

Shabman, Leonard A. and Michael K. Bertelsen, The Develo ment Value of
Natural Coastal Wetlands: A Framework for Analys1s of Res1dential
Values. Sea Grant Research Report No. A.E. 35, Virginia Tech,
Blacksburg, VA.

 

Shaffer, James 0., 1979A, ”Observations on the Political Economics of
Regulations.” American Journal of Agricultural Economics, Vol. 61,
No. 4 (Part 2), pp. 721-731, November 1979.

., 19793, "Food System Organization and Perfonnance:
Toward a Conceptual Framework. " American Journal of Agricultural
Economics, Vol. 62, No. 2, pp. 310- 318, May 1980.

 

. "On Institutional Obsolescence and Innovation—
Background for Professional Dialogue on Public Policy.” American
Journal of Agricultural Economics, Vol. 51, No. 2, pp. 245-267,
May 1969

Small, Kenneth A. "Air Pollution and Property Values: Further Comment.”
Review of Economics and Statistics, Vol. 57, pp. 105-107, 1974.

Steiner, Peter 0. ”Choosing Between Alternative Investments in the
Water Resources Field." American Economic Review, Vol. 49, No. 5,
pp. 893-916, 1959.

 

Strotz, Robert H. "The Use of Land Value Changes to Measure the Welfare
Benefits of Land Improvements," in Joseph E. Haring, ed., The New

Economics of Regulated Industries, Occidental College, Los Angeles,
CA, 1968.

Stull, William J. "Community, Environment, Zoning and the Value of
Single-Family Homes." Journal of Law and Economics, Vol. 18, No. 2,
pp. 535—558, 1975.

Tilton, Donald L., Robert H. Kadlec, and Benedict R. Schuegler, The
Ecology and Values of Michigan' 5 Coastal Wetlands. U. S. Fish _and
Wildlife Service, Twin Cities, MN, 1978.

United States Environmental Protection Agency, "Wetlands: Protecting a
Fragile Environment." EPA Environment Midwest, pp. 10-16, March
1980.

 

 

180

U.S. Water Resources Council, Principles and Standards for Planning Water

 

and Related Land Resources. Federal Register, Vol. 38, No. 174,
pp. 24778—24851, September 10, 1973.

 

Walker, Richard A. ”Wetlands Preservation and Management in Chesapeake
Bay: The Role of Science in Natural Resources Policy." Coastal
Zone Management Journal, Vol. 1, No. 1, pp. 75-101, 1973.

Whigham, Dennis F. and Suzanne E. Bayley, ”Nutrient Dynamics in Fresh
Water Wetlands." Wetland Functions and Values: The State of Our
Understanding. P. E. Greeson, J. R. Clark and J. E. Clark, eds.,
American Water Resources Association, Minneapolis, MN, pp. 468-
478, 1979.

Van der Valk, Arnold G., Craig B. Davis, James L. Baker, and Craig E.
Beer, "Natural Fresh Water Wetlands as Nitrogen and Phosphorus Traps
for Land Runoff.” Wetland Functions and Values: The State of Our
Understandin . P. E. Greeson, J. R. Clark and J. E. Clark, eds.,
American Water Resource Association, Minneapolis, MN, pp. 457—467,
1979.

 

   

 

 

 

 

 

 

 

 

   

"I111111111111111