Improving the oil production forecast in Alaska. Bill Barron, DNR and Bruce Tangeman, DOR, Alaska OIl

1. State of Alaska
Department of Revenue
December 6, 2012
1

2. What is the problem?
In developing the forecast DOR Forecasts
the Department & its 1,200
consultant rely on input
from industry to forecast 1,100
production.
1,000
While this method works 900
BOPD
relatively well in the near
term it has consistently 800
been optimistic in the
700
long-term.
600
The average forecast error
for six to ten years in the 500
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
future has been 40%-60%. Actual FALL 01 FALL 03 FALL 05 FALL 06
2

3. The Department’s Approach
The Production Forecast has been based on three
tranches of future production:
ANS Production Decline
Currently Producing: 1000000
Oil from wells that are in production
and following typical reservoir
engineering optimization without
major investment.
Under Development (UD):
•Oil from projects that will add
incremental oil to existing fields or will
bring new fields into production.
•Project must have senior management
approval and be allocated funds in the
companies budget.
Under Evaluation (UE):
•Oil from projects that are likely to
occur in the future, but have not meet
the requirements of the previous Bounds of reasonable expectations
category. for future production were developed
•Requires that oil reserves are known
and recovery is technically possible based on historical production
with current technology.
Under Development +
Under Evaluation = “New Oil” 100000
2020
2021
2022
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
***These definitions are not equivalent to those used by the Society of Petroleum Engineers (SPE) or
Securities & Exchange Commission (SEC) and should not be used as such***
3

4. Accounting for the Risks Appropriately
 Currently Producing oil was not risked because it is grounded in
engineering principles.
 The New Oil portion of the Forecast was adjusted using a
scenario modeling method.
 Categories
 Budgetary
 Technical
 Confidence decreasing with time
 Higher confidence with near term, 1-3 years, than in out-year plans.
 Reasonable to expect greater uncertainty in both under-
development and under-evaluation tranches.
 Recognizing the impact of the known-knowns, known-unknowns and
unknown-unknowns.
4

5. Testing the Traditional Forecast
Forecasted ANS Production
1000000
100000
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022
5

6. Applying the Refined Methodology
ANS Production Forecast
1000000
100000
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022
6

7. Summary
 Prior forecasts have over-estimated long-term (6 years out)
production by significant margins (40-60%).
 The refined methodology is based on statistical analysis and
tested against past performance.
 Reasonable and Prudent
 The refined methodology does not impact near term revenues.
 Provides a more prudent and reasonable forecast for the
purposes on long-term planning.
 Alaska still holds substantial discovered and undiscovered
resources that may provide additional production in the future.
 Example – heavy and shale oil are not included in the forecast.
7