Sei sulla pagina 1di 25

Static Models, Dynamic

Processes and the Tongass


Land Management Plan

Guy Robertson

Research & Development


US Forest Service
Washington Office
Overview

• Two Central Arguments


– Static, deterministic (and complex) models
don’t work in predicting futures for dynamic,
stochastic systems
– Adaptive management approaches are very
difficult to implement in highly contentious
planning environments
Overview
• Strategy
Comparison of ex ante predictions and ex post
outcomes associated with the Tongass Land
Management Plan in Southeast Alaska
• General planning process
• Economic Impact Analysis
• Predicted harvest levels
Conclude with discussion of new FS planning
rule
Tongass Land Management Plan
• Part of a recurrent forest planning process
stipulated by the NFMA
• Subject to NEPA requiring an EIS
• Anticipated a reduction of timber harvest
• Highly contentious
Framed as a classic conflict between jobs and
the environment
Technocratic Planning Model
Formal Process

Within Agency

IDT Produces FS Decision


Decision
NEPA Document Maker

“Science” Public
Participation
Deterministic Planning Model
Nested within technocratic model
Planning Direct Economic Indirect Economic
Alternatives Impacts Impacts
(IMPLAN)

LUDS Harvest
Polygons Scheduling Outputs
On map (FORPLAN) •Timber
•Fish Decision
•Recreation
•Wilderness
Maker
•Etc.
Other Estimation Techniques
(e.g. deer habitat model) Other Impacts
Social Choice Planning Model

Broader Social Arena


•Politics
Formal Process •Press Policy Outcome
•Public Opinion
•Courts
TLMP Preferred Alternative:
Called for a 200 million board foot (MMbf)
reduction in harvest
(down from previous average of 450 MMbf)

Projected Employment Impacts:


per @ -200
MMbf MMbf
Direct Employment 3.03 -607
Total employment (Multiplier = 2.93) 8.89 -1,778
What Actually Happened:
Timber Harvest
120

100
Million Board feet

80

Offered
60 Sold
Harvested

40

20

0
2001 2002 2003 2004 2005

Markets didn’t cooperate…


What Actually Happened:
Direct Employment

4,000

Plan “implemented”
3,500

3,000

2,500
Jobs

2,000

1,500
1,052 Direct jobs
1,000
lost since 1997

500

0
82

83

86

87

90

91

94

95

98

99

03

04
02
00
84

85

88

89

92

93

96

97

01

05
19

19

19

19

19

19

19

19

19

19

20

20

20
19

19

19

19

19

19

19

19

20

20

20
What about the indirect and
Induced employment impacts?
@ 2.93 total jobs per direct job, impacts
would be on the order of:
1,052 x 2.93 = 3,082 jobs
52,000

51,000

50,000
Jobs

49,000

48,000

47,000

46,000
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Total Employment in SE Alaska


In Timber Production Counties?
19,000

18,000

17,000

16,000

15,000

14,000

13,000

12,000

11,000

10,000
80

86

88

90

92

94

96

00

02

04
82

84

98
19

19

19

19

19

19

19

19

19

19

20

20

20
Total Employment in Ketchikan, Prince of Wales, and Wrangel
Yakutat
250

200
Basic
Employees

150
Non-Basic
100

50
Evidence of
Multipliers
0
Haines
900
800

…?
700
600 Basic
Employees

500
Non-Basic
400
300
200
100
0

Kake
250

200

Basic
“Basic” and
Employees

150

100
Non-Basic “Non-Basic”
50
employment in
selected
0
communities
82.1

83.2

84.3

85.4

87.1

88.2

89.3

90.4

92.1

93.2

94.3

95.4

Year & Quarter


300 Angoon 100 Gustavus

200 50
0
100
-50
0

Evidence of
-100
-100 -150
-200 -200

Multipliers
82 84 86 88 90 92 94 96 82 84 86 88 90 92 94 96

NONBASIC BASIC NONBASIC BASIC

300 Hollis 150 Kake

…?
200 100
100
50
0
0
-100
-50
-200
-300 -100
-400 82 84 86 88 90 92 94 96
-150 82 84 86 88 90 92 94 96

NONBASIC BASIC NONBASIC BASIC

400 Wrangell 120 Yakutat

200
80 “Basic” and “Non-
40 Basic”
0
0 employment in
-200
-40 selected
-400 82 84 86 88 90 92 94 96
-80 82 84 86 88 90 92 94 96
communities (first
NONBASIC BASIC NONBASIC BASIC
differenced)
Evidence of Multipliers
…?
2.5
2
1.5
1
0.5
0
-0.5
-1
-1.5
-2
-2.5
Sitka

Petersburg
Thorne Bay
Yakutat

Kake
Wrangell

Ketchikan

Juneau

Gustavus
Metlakatla

Hydaburg

Hoonah

Hollis
Haines
Angoon

Regression Estimates of Multipliers with


+/- 2 standard deviations
Evidence of Multipliers
…?
No

But why?
– Leakage
– Labor inelasticity
– Static models can’t predict dynamic
systems…not even as approximations
So was the TLMP a Failure?

No
– Public information document
– Reference to keep the debaters honest
– Focus for social and legal debate
• Current conditions
• Projections
But not a success in the narrow technocratic
decision model sense
So was the TLMP a Failure?

Also numerous “setbacks”


– 1999 New ROD released to address 33 TLMP Appeals.
– 2001 1999 ROD vacated; 1997 ROD reinstated.
– 2003 February 2003: Supplemental Environmental Impact Statement –
Roadless Area Evaluation for Wilderness Recommendations Record of
Decision.
– 2005 Ninth Circuit Court claims inadequacies in NEPA procedures
requires additional adjustments and updates of the Forest Plan (officially
called a Forest Plan Amendment).
– 2007 The Draft Environmental Impact Statement and proposed Forest
Plan Amendment became available for public review

Reaffirms broader social arena as locus of decision making


So what to do?

Adaptive Management
– FS buzz word for at least ten years
– Monitoring plan required in all planning exercises
– Appx. $400 million for inventory and monitoring activities

But, in response to potential legal challenges


– “Bullet proof” the plan
– Check off the monitoring report as an afterthought
This is not a very flexible approach
New FS Planning Rule

Vision Strategy Design Criteria


Forest Level Project Level Forest Level
Long-term Short-term Mid-scale
Collaborative Collaborative Technical/Collaborative

Combined with EMS and third party monitoring

Internalizes social choice dynamics through collaboration


New FS Planning Rule

Aims to provide a flexible planning


approach that avoids conflict and legal
challenges
FS argues that the “vision stage” is not subject to NEPA (EIS) as it does
not involve actions on the ground

Unclear if and how conflict will be arbitrated

However, the Rule itself was successfully challenged in


9th circuit owing to lack of EIS
Conclusion

• Technocratic model is out of touch with


reality
– Failure to adequately predict
– Decision locus not with the agency
• Communication and information flow more
important than technique
– Fancy models and false precision
Conclusion

• Adaptive management problematic in


conflict environment
– From an agency perspective at least
– From broader social perspective system may
be adaptive, though cumbersome and
expensive
• Anticipating a post-conflict future for public
forest management in US

Potrebbero piacerti anche