Title: Improving the policy process
1Improving the policy process
PAHO Leaders 2006 Disaster risk reduction,
mitigation, preparedness and response
- Pan American Health Organization
- Regional Center for Disaster Risk Reduction, UWI
- CDERA
-
- Thursday 23rd November 2006
- Sunset Jamaica Grande Resort, Ocho Rios
Professor Anthony Clayton University of the West
Indies
15/05/2020
2Problems, problems
- Problem why do so many government policies fail?
- Common reasons failure to understand the task,
unhelpful misperceptions, assumptions and
politics, policy conflicts, resulting in
incoherent priorities unachievable missions. - Examples
- Why do countries get poorer when aid flows
increase? - Why do we negotiate for aid programs that damage
our economy? - Why do measures to protect jobs increase
unemployment? - Why do government agencies undermine each other?
- Why cant we solve the drug problem?
3Example War
- The nature of war has changed
- Wars were formerly between nations the majority
of conflicts today involve civil
wars/insurgencies. War today is more likely to
involve infiltration rather than invasion. - Wars are increasingly asymmetrical, between
security forces and unconventional irregular
forces. - In traditional warfare, most of the fatalities
were soldiers. Today, most of the casualties are
civilians. - The nature of war has changed fundamentally more
significant changes are expected over the next
10-20 years. - The way in which the issues are perceived by the
general public (taxpayers, voters, recruits) has
not kept pace. So there is an increasing
disconnect. - Even more important, the way in which the issues
are perceived by the professionals has not kept
pace
4Wargame exposed gaping hole in Pentagon strategy
Julian Borger, The Guardian Tuesday September 25,
2001
- When terror came out of a clear blue sky on
September 11, some of the Pentagon's top brass
were given a jolting reminder of a wargame they
played in 1997. In the game, the US was pitted
against a zealous, decentralised terror
organisation very like al-Qaida, and the US lost.
Each time the US forces thought they had scored a
blow against the terror organisation, it would
regenerate itself to strike on another front. It
emerged from the campaign more or less unscathed.
- That was four years ago, but some of the military
strategists who organized the game believe the
Pentagon failed to learn the lessons from the
wargame, and said yesterday that the US military
was still a long way from readiness to fight the
adversary it now faced.
5The Pentagon still fighting the wrong war
- According to Doug Johnson, a strategic studies
professor involved in the 1997 exercise, the army
lost because it didn't want to play that game.
The army brass, Dr Johnson said, were intent on
fighting a variation of a war against large tank
armies on the central plains of somewhere. At
one point, the Pentagon officers involved became
so frustrated with their elusive opponents that
they asked for the game organizers to have a
friendly government's armoured battalion defect
to the other side. They did it to give someone
to blast, Dr Johnson said. Everyone went away
feeling viscerally satisfied. - As a result, they missed the point. The terror
organisation still had most of its cells in
place, and a functioning financial network. - Within the contours of that particular game, the
American forces and their allies simply weren't
configured to deal with an enemy like the one we
created, said Steven Metz, head of the Army War
College's regional strategy and planning
department.
6Lessons?
- This illustrates three points in policy analysis
- Distinguish between perception and reality.
- Be wary of the Conventional Wisdom of the
Dominant Group (COWDUNG), as a failure on a core
assumption is likely to be fatal. - The importance of seeing both the current
position and the underlying context or trend.
7Making better policy
- Question how can we increase the chances of
success? - Answer a more systematic approach, to help us
- Control for preconceived ideas and assumptions
- Check the facts, get the problem into focus
- Identify our options
- For each option, identify costs and trade-offs
- Agree our priorities
- Build the necessary consensus about the way
forward - Implement!
- Monitor, change tactics if required, stay focused
on the goal
8Bardachs Eightfold Path lists the main steps
- Define the problem. What are we trying to
achieve? - Assemble the evidence. What are the facts?
- Construct policy alternatives. What can we do?
- Select the criteria. How are we going to decide
which is the best policy? - Project the outcomes of each policy option. What
would happen if we tried this? - Assess the trade-offs. How much will this cost?
What are the chances of success? Could this
tactic create some other problem? - Decide. What are you going to recommend to the
Minister? - Present. Set out the problem, your analysis and
your recommendation.
93) Constructing policy alternatives
- Must be evidence-based (from stage 2).
- Take into account a range of expert advice and
stakeholder views. - Represent the main bodies of opinion.
- Help to bring issues into focus (usually achieved
by making sure that alternatives are sufficiently
diverse). - Example 1 Swedish defence policy
- Example 2 Go Big, Go Long, Go Home.
104) Select the criteria
- What is important? Is it
- Results at any cost
- Results, provided that the process is not too
expensive - Results, provided that they can be achieved
without upsetting anyone with a lot of political
influence - Not to change anything, but to make it look as
though we are taking the problem seriously - So does the proposed solution have to be
- The solution that we think has the best chance of
success? - The cheapest/quickest/most legal solution
available? - The most politically-acceptable way forward?
- The one that will make us look good?
115) Project outcomes for each option
- This is most technically difficult step, because
- We have to be realistic, rather than optimistic
or pessimistic - We have to resolve very significant uncertainty
into a clear decision, on which much will depend. - It is about the future, NOT the present
-
- Key questions what would happen if we did this?
- Have we tried anything like this before? What
happened that time? Why? - Has anyone else tried anything like this? What
happened to them? Were the conditions similar? - How will people react if we do this? How will the
community respond? Will there be political
problems? How will the different interest groups
react? Could there be legal challenges?
12Estimating impacts and costs
- Example the cost of crime control in Jamaica
- For J180m we can reduce homicides by 5.
- If 5 80 homicides, cost of reduction
2.25m/homicide. - What are the chances of success? 90? 50? 10?
1? - What are the risks of unwanted side effects?
- How does this compare against e.g. investment in
schools? - Bearing in mind that
- Job security increases unemployment
- Safer drivers take more risks
- Free zones encourage relocation, not job-creation
136) Dealing with trade-offs
- One option is clearly the best (rare)
- Each option has a different combination of costs
and benefits (the usual situation)
Option Cost Chance of success Risks
A Medium High High
B Low Low Medium
C High Medium Low
14Aggregated vs disaggregated decision-making
- Aggregated resolve everything into a single
numeraire (usually cash). - Disaggregated use matrices, maps and diagrams to
keep more dimensions in view. - There are pros and cons for each approach.
15Aggregated
Low 0-3 Option make negative Medium
4-6 factors (e.g. cost, risk) High 7-10 negative
numbers.
Option Factor A - cost Factor B prospects Weighting 2 Factor C - risks Total
Plan A (Medium) -5 (High) 8 16 (High) -8 3
Plan B (Low) -2 (Low) 3 6 (Medium) -4 0
Plan C (High) -7 (Medium) 6 12 (Low) -3 2
This one!
16Probability/impact matrix
- The conclusions from a policy exercise usually
have to be absorbed into a government (or
business) agenda that is already crowded. It is
important to have greater clarity about future
problems and potential solutions, but that does
not remove the need to make the large number of
day-to-day decisions involved in managing a
Ministry or government agency. - So it is important to have priorities for action.
- A probability impact matrix is a way of
organizing these priorities. It is similar to the
triage used by military doctors when dealing with
incoming casualties.
17Probability/impact matrix
Low impact High impact
Low probability
High probability
Ignore
Monitor carefully
Risk of critical failure!
Low priority
Top priority
18High impact, low probability (we thought)
- It is sensible to allocate most time and
resources to high impact, high probability
events, but it is also important to monitor high
impact, low probability events - The US Federal Emergency Management Agency knew
that New Orleans was potentially vulnerable to a
hurricane, and had identified this as one of the
three worst disasters that could befall the
United States. This was, however, seen by the
administration as a low probability event. - On the 28th29th August 2005, Hurricane Katrina
resulted in a 28-foot storm surge as well as
torrential rain, the latter raised the height of
Lake Pontchartrain by 7.6 feet, and the
combination overwhelmed the levees that protected
the city of New Orleans. About 80 of the city,
which is on average about 6 feet below sea level,
was then flooded.
- Unprepared. The Washington Post. Monday,
September 5th, 2005
197) Decide
- Are you convinced?
- Are your colleagues convinced?
- Are the external experts, independent advisors,
stakeholders convinced?
208) Final presentation
- State the following
- The problem
- The analysis of trends, causes
- Options, intervention points
- Benefits, costs, trade-offs and risks
- Recommendation
21Project management tools
- There are a range of useful tools, including
- Project cycle management
- LogFrame planning
- Critical Path Analysis
- PERT analysis
- Gantt charts
- These can help to
- Identify the important sub-tasks.
- Identify the dependencies between sub-tasks.
- Organize the dependent sub-tasks into the
appropriate sequence. - Identify potential vulnerabilities.
- Identify the minimum time required to complete a
project. - Identify where resources can be optimally
allocated.
22Project cycle management (1)
- Most projects can be divided into a 6 phase
project cycle - Defining the theme. This is the choice of area,
usually a list of key problems that require a
policy intervention (typically scoring high on
risk, public concern, or assessed chances of
success). - Project identification. This is the initial
formulation of ideas for the actual project,
including objectives, expected results and a list
of the actions to be taken. The goal at this
stage is to work out whether it is worth going
ahead with a feasibility/pilot study. If the
answer is yes, then the next step is usually to
draw up the terms of reference and undertake the
study. - Formulation. This involves looking at the results
of the feasibility or pilot study and specifying
the project itself. Objectives, expected results
and a list of the actions to be taken must now be
set down in precise detail. These must then be
checked back against the theme defined earlier.
If the project is going to require external
funding, it is at this stage that you decide
whether or not to draw up a bid for funds.
23Project cycle management (2)
- Financing. For those projects that require
external funding this stage involves drafting
the funding proposal, bidding for and securing
funds, negotiating and signing the contracts. - Implementation. This involves executing the
project, monitoring progress and achieving the
results. - Evaluation. This stage involves analyzing the
results and assessing the impact of the project
both during and after its implementation. The
goal is to identify any lessons to be learnt that
could help with other projects, both now and in
the future. Projects that require external
funding are often phased, with the release of
funding for each stage conditional on the
completion of a full evaluation of the previous
stage.
24LogFrame risk reduction
- What is LogFrame?
- Logical Framework (LogFrame) planning involves
analyzing and setting out, in a systematic and
logical way, the objectives of a project and any
causal relationships between the objectives of a
project. - It involves working towards these objectives,
monitoring progress and checking that the
objectives have been achieved. It also involves
establishing the extent to which the success of
the project depends on factors that cannot be
controlled, and the extent to which the project
is therefore exposed to risk. - How LogFrame help?
- By helping us to think methodically and identify
all the factors essential to the success of the
project. - By encouraging us to check and test our ideas,
monitor progress, identify problems and take
remedial action while there is still time to save
the project.
25LogFrame matrix
The activities can start if the preconditions are
met the activities will lead to results if the
assumptions are met the results will accomplish
the project purpose if the assumptions are met
achieving the projects purpose will help to
achieve the overall objectives if the assumptions
are met.
Intervention logic Verifiable indicators Sources of verification Assumptions
Overall objectives (the contribution)
Project purpose (the point)
Results (the achievement)
Activities (the actions) Means Costs
Preconditions Start
26Critical Path Analysis (1)
- Critical Path Analysis and the related PERT model
were developed in the 1950s to manage military
projects, but are now more generally used to
manage any particularly large, complex project.
Another related tool, the Gantt chart, was
developed three decades earlier. These tools
involve listing all the sub-tasks in a project,
then organizing them into two groups - Sequential the first group consists of those
tasks that have to be completed in sequence,
because each stage depends on the one before.
When building a factory, for example, the
foundation must be finished before the
load-bearing walls go up, and the walls must be
ready before the roof can be fitted. - Parallel the second group consists of those
tasks that do not depend on the completion of
other tasks. These can therefore be completed in
parallel, i.e. at the same time as other tasks.
In the same factory building project, for
example, the contractor might decide that the
tarmac for the car park can be laid at any time
this does not depend on progress with the
foundations, walls or floors of the main building.
27Critical Path Analysis (2)
- The tasks are then represented in a diagram which
shows the flow of events. The critical path is
the line through the series of sequential events.
This shows the minimum amount of time needed to
complete the project as a whole. This also shows
where the project might be vulnerable, because
any failure on the critical path will always have
implications for either the timetable (the
completion will be delayed) or the budget (we
will have to hire more people to get this phase
finished on schedule). The key points along the
critical path usually serve as the decision
points for the project. - This process shows where additional resources
would have the most effect. Additional
expenditure on a critical path event can shorten
the amount of time required, and help to get a
late project back on schedule. Hiring more
bricklayers, for example, can help to get the
walls finished earlier, thus making it possible
to get the roof fitted. Additional expenditure on
a parallel task, however, such as bringing in
another roller to level the car park, will not
help to shorten the timetable. This reveals how
resources can be reallocated from parallel tasks
to sequential tasks in order to speed up a
project. This sort of tactic is sometimes
referred to as a crash action programme.
28Program Evaluation Review Technique (PERT)
- PERT is a form of Critical Path Analysis which
also corrects for the fact that most people
underestimate how long each task will take, while
some people will overestimate the time required
in order to inflate their bonuses. PERT is
calculated by estimating the shortest possible
time each task will take, the longest likely time
each task could take, and the most likely amount
of time that each task will actually take. In
effect, PERT uses a band of values, with a top
and bottom end and a most likely value, as
opposed to a single value. - This band is then resolved into a single value,
usually in the following formula shortest time
4 the most likely time the longest time/6. So
if, for example, the shortest time was 2 days,
the longest time 8 days, and the most likely time
4 days, that would give 2 (44) 8 26/6
4.3 days. - This final value is then used instead of the
shortest time value of 2 days or the longest
value of 8 days, thus correcting for
over-optimism or under-bidding.
29Visual tools
- Critical path, PERT and Gantt charts (Gantt
charts were developed by Henry Gantt in 1917) are
usually shown as horizontal bar charts that show
the important tasks over time. - These are useful visual tools for keeping track
of projects.
30How to make Critical Path, PERT or Gantt charts
- Make a list of all tasks/activities involved in
the project. - Put them into sequential order.
- Estimate the time it will take to complete each
task and put the time next to that task on the
list. - Readjust the sequence of tasks as necessary.
- Determine who is responsible for each task on the
list and put their name next to that task. - Label the chart across the top by day/week/month.
- Label the chart along the left side with all the
tasks. - Draw horizontal bars for each task beginning at
the start date for that task and ending with the
completion date for that task.
31Example
Critical path (sequential) Non-Critical
(non-sequential)
Activity i/c Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Prepare proposal Fred
Secure budget John
Recruit new staff Mary
Train new staff Joan
Assess existing staff Dave
Maintain equipment Fred
Maintain premises Bill
325) Project outcomes for each option
- This is most technically difficult step, because
- We have to be realistic, rather than optimistic
or pessimistic - We have to resolve very significant uncertainty
into a clear decision, on which much will depend. - It is about the future, NOT the present
So how do we plan for the future?
33The problem with forecasts
- Sophisticated, highly-paid economists have
successfully predicted seven of the last three
global recessions (various sources). -
- Never make forecasts especially about the
future - - Sam Goldwyn
34Why do forecasts fail?
- 1) Trends change. We start from the current
position, identify and extrapolate trends. We
tend to assume that these will continue but
eventually they dont. Competition intensifies,
markets saturate and preferences change. - 2) Parameters change. Market trends are shaped by
underlying social and economic factors e.g.
demographics, productivity growth and
technological development. These structural
factors usually change slowly but external
shocks or new disruptive technologies can
represent discontinuities (or tipping points)
that precipitate more rapid change and
restructure markets.
- Trend change is (in principle) predictable.but
it is difficult to anticipate discontinuities.
35Can we do better?
- Conventional forecasting results become less
reliable at times of significant, rapid and
pervasive change - We are living in an unprecedented era of
accelerating technological, economic and social
change, driven by - New technologies, such as informatics,
biotechnology, and nanotechnology - Global trade liberalisation, which could increase
contestable world output from 20 of the total to
80 of the total by 2030 - The re-drawing of world geopolitical parameters
- Changing pattern of resource demand,
environmental impacts - Are there more reliable decision/planning tools
for times of change?
36What is foresighting?
- Foresighting is a tool used
- to inform decision-making
- to help people anticipate future opportunities
and avoid problems - Foresighting involves
- envisioning a range of possible future scenarios,
- testing these rigorously, then -
- back-casting to present day, and mapping out the
steps needed to achieve the preferred
scenario/avoid the worst-case scenario. - The foresight process itself is important,
because - it clarifies and challenges assumptions
- it encourages shared visions and flexible
thinking - it creates new knowledge networks
37Changing parameters climate change
38Earth's temperature is dangerously high - NASA
- Researchers at Nasa's Goddard Institute for
Space Studies said that Earth's temperature was
now reaching its highest level in a million
years. Dr James Hansen, who led the study, said
further global warming of just 1C could lead to
big changes to the planet. If warming is kept
less than that, effects of global warming may be
relatively manageable, he said. But if further
global warming reaches 2 or 3C, the Earth may
become a different planet to the one we know
now. The last time it was that warm was in the
middle Pliocene, about 3m years ago, when sea
level was about 25 meters (80 feet) higher than
today. - The study showed that there was already a threat
of more extreme weather like the strong El Niños
in 1983 and 1998, when many countries around the
world had devastating floods and tornadoes.
Adapted from Hilary Osborne Tuesday September 26
2006 The Guardian
39Hurricane Katrina, 2005, S E of New Orleans
40The dispossessed.
Flooding in Bangladesh
41(No Transcript)
42Is this a solvable problem?
Previous environmental treaties have had partial
success The Montreal Protocol, which limits CFC
emissions. The Basle Convention, which controls
trans-boundary shipments of hazardous waste. But
these are relatively solvable problems compared
to energy use carbon emissions derive from the
use of our primary energy sources.
43Kyoto redundant before ratified
- The US, the largest source of carbon emissions,
has not ratified the protocol, partly because it
imposes no limits on the gases produced by
developing countries. - China, which is now the worlds biggest consumer
of coal and second biggest consumer of oil, emits
almost as much carbon as the 25 members of the EU
combined, and will shortly overtake them to
become the worlds second largest source of
carbon emissions, is exempt. - As a result of these non-ratifications and
exemptions, UN projections indicate that the
treaty will reduce the currently projected rise
in average surface temperature of 1.4 to 5.8C by
2100 by just 0.1.
44Methane release
- potential disaster?
- There are naturally-occurring greenhouse gases,
mostly methane, trapped in ice-like structures
called clathrates. Most of these are trapped in
cold sediments and Arctic tundra. - There is 400 gigatons of methane currently
trapped in frozen arctic tundra. If the
temperature gets too high, and the tundra
defrosts, this methane will be released. Methane
is gt20 times more efficient than CO² as a
greenhouse gas, so this could cause runaway
climate change. - Similar events have happened before. The largest
previous release of methane happened at the
Permian-Triassic boundary event, about 250
million years ago, when 95 of extant species
were destroyed. It took 20 - 30 million years for
rudimentary coral reefs to re-establish and
forests to re-grow in some areas it took gt100
million years for ecosystems to reach similar
levels of diversity.
45Western Siberia in 2005...thawed for the first
time in 11,000 years
46A Stern warning (part 1)
- Key points in the report written by Sir Nicholas
Stern for the UK government, published on Monday
30th October - CO² and temperature rise
- Carbon emissions have raised global temperatures
by 0.5C. - With BAU, there is gt75 chance that global
temperatures will rise by 2-3C over the next 50
years. There is a 50 chance that global
temperatures could rise by 5C. - Environmental impact
- Melting glaciers will increase flood risk, then
drought. - Crop yields will decline, particularly in Africa.
- Rising sea levels could displace 200 million
people. - Up to 40 of species could become extinct.
- There will be more frequent extreme weather
patterns.
47A Stern warning (part 2)
- Economic impact
- A rise of 2-3C could reduce global GDP by 3.
- A rise of 5C could cost up to 10 of global GDP.
The poorest countries would lose
disproportionately more. - Worst case scenario the global economy could
shrink by 20 - permanently. - Cost of remedial action
- Controlling this risk would require stabilizing
emissions within the next 20 years then reducing
by 1-3 pa. The transition to a low-carbon
economy would cost 1 of GDP, mostly one-off
expenditure (e.g. investment in low-carbon
technologies). - Conclusion
- A one-off investment of 1 could avert a
permanent reduction in annual income of 5-20.
48(No Transcript)
49The real problemas always
is people
50Surging demand
- Transport still only accounts for 14 global
emissions, less than power generation and
land-use. - However, air travel is the most rapidly-growing
source of carbon emissions. - This growth in the demand for global transport
(driven by falling prices and increasing demand)
is now seen as one of the most intractable
problems in slowing the rate of climate change.
51Growth in air travel
- Air traffic has been expanding at nearly 250 of
average economic growth rates since 1959. - The current UK Government's Aviation White Paper
notes that aviation has increased fivefold over
the last 30 years, and predicts that UK passenger
numbers will more than double from 180 million to
475 million over the next 25 years.
52Carbon-intensive activity
- Air travel is not only the most rapidly-growing
source of carbon emissions, it is also very
energy-intensive. For example, a Boeing 747 burns
about 5 gallons per mile, so a London to New York
flight (3,471 miles) requires some 17,355
gallons. - Air travel relies on carbon-based fuel, so it is
a carbon-intensive activity more so than other
forms of transport. For example, a short flight
(under 500km, e.g. London to Amsterdam) generates
0.17kg of carbon/passenger/kilometre, compared
with 0.14 kg/km by car, 0.052 kg/km for rail and
0.047 kg/km by ship. So flying produces over
three times more carbon/kilometre than rail, and
over three-and-a-half times more than travelling
by ship. - As a result, a return flight between the UK and
Australia produces about 3 tonnes of carbon per
person (for comparison, an average car emits
about 6 tonnes of carbon per annum).
53Aggregate carbon emissions
- The world fleet now consists of about 16,000
commercial jet aircraft. These generate over 600m
tonnes of CO² per year. - This means that aviation generates now nearly as
much CO² as all human activities in Africa.
54Consuming the carbon allowance
- In September 2005 the UK's Tyndall Centre for
Climate Change calculated that all householders,
motorists and businesses in the UK would have to
reduce their CO² emissions to zero if the
aviation industry was to be incorporated into the
UK Government climate change targets for 2050. - In other words, the entire UK economy would have
to emit no carbon at all, because the airline
industry would be emitting so much that it would
consume the UKs entire carbon allowance. The
same equation is true of most other EU member
states. This is clearly impossible.
55Technological options?
- There have been significant improvements in
aircraft and engine technology. The new
generation of planes (like the Airbus A380 and
the Boeing 7E7) are partly made of carbon
composites, rather than metal, therefore offer
significantly better fuel efficiency and reduced
emissions per passenger. - But the IPCC point out that these gains will be
erased by the projected growth in demand, which
means that total fuel consumption and emissions
of e.g. carbon dioxide, water vapour, nitric
oxide, nitrogen dioxide and sulphur dioxide will
continue to rise. - It is not (yet) possible to operate aircraft on
biofuels (but do we have enough land anyway?)
56Scarcity is dynamic
- Scarcity reflects four main factors
- Total physical quantity in available
form/location. - Technology required to find, extract, process and
transport the resource. - Pattern of demand/use (also partly determined by
available technology). - Economics (anticipated) demand, supply,
competition, alternatives substitutes. - These four factors determine energy
resource-use efficiency. - So calculations of sustainability must take into
account a number of factors, including (a) total
stock, location and accessibility (b) available
technology, (c) economics (d) market and
technological trends, (e) development trends (f)
demand growth and so on.
57Solutions
- Faster technological advance, including both
incremental gains and more radical substitutions. - Faster progress in offset (flanking)
technologies. - Demand-rationing, via price.
- Command-and-control, via regulation.
- Disaster preparedness, zoning, planning,
exercises and mitigation.
58Conclusions
- We need a robust process for making, implementing
and monitoring policy. - This has to be based on strategic/scenario
planning and risk mapping. - We need to search for the optimal (attainable,
achievable or affordable) solution this might
lie in another discipline or jurisdiction.
59Thank you !