Extractive Industry Indicators

1
Extractive Industry Indicators

• Data on EITI and non-EITI Mineral Exporters
• Susan Aaronson, GWU and Carla Winston, USIP

2
Overview part one

• This power point serves to summarize our
  analysis of the data we used to compare EITI and
  non EITI mineral/energy exporters in the period
  2000-2007 (Data was not always available for each
  country for every year.) We looked at the means
  and medians, and used percentiles to describe
  differences in performance over time. The data
  sets compare the 26 EITI with the 25
  hydrocarbon-rich countries and mineral rich
  countries monitored by the IMF in the period
  2000-2006. The purpose of this data review is
  not to assert causality, but rather to gain
  greater understanding of how a decision to adopt
  EITI may affect the behavior and performance on
  various indices of energy/mineral rich states.1
  
• Because every country is unique, the data
  shows significant variance among energy/mineral
  states. That fact, as well as the short time
  period of EITI implementation, may explain why
  differences in development outcomes (such as
  reduction of corruption etc ) dont always favor
  greater progress among EITI countries vs.
  non-EITI.2 Time and further research will tell
  if EITI can help these countries to use
  energy/mineral revenues more wisely.

3
What countries did we examine?

• The EITI 26 were subdivided into groups based on
  their development level (high medium and low) and
  the duration and effectiveness of their adherence
  to EITI, based on their implementation of the
  first 4 Indicators of EITI).
• The comparison group did not include those
  countries which have high development rank in the
  UNDP Human Development index rankings (Bahrain,
  Brunei, Kuwait, Mexico, Norway, Oman, Qatar,
  Saudi Arabia, United Arab Emirates, and Chile).
  We also excluded Iraq from the group because of
  the war and uneven data.

4
Chart of Hypotheses
5
Summary of Hypotheses
6
Doing Business Rankings

• These charts represent the change in rankings on
  the IBRDs Doing Business index for the years
  2006 and 2007. The Doing Business Index is a
  Guide for evaluating regulations that directly
  impact economic growth, downloading underlying
  laws, making cross-country comparisons, and
  identifying good practice reforms. The first
  rankings were in 2006 (using 2005 data) and
  covered 175 countries.
• www.doingbusiness.org/.

7
Individual country changes are in gray averages
in green, and medians in blue. The Y-axis
represents improvement in rankings by overall
points, not percentages. Some countries improved,
others didnt. 11 of the 26 EITI sample-- 42
were able to improve the economic climate to
foster economic growth and 5 of the 11 made
significant improvements. The average
improvement was 9.72 points in the Doing Business
ranking.
8
In general, the non-EITI countries have fared
worse than the EITI countries in improving the
business climate. Interestingly, the median
change for medium-developed countries is
positive, while the median change for
low-developed countries is quite negative.
9
Comparing the countries based on development, the
average ranking for all EITI groupings has
increased the average ranking for all non-EITI
countries has decreased.
10
However, when we compared the medians the
non-EITI countries improved, and the EITI
countries did not. EITI median development stayed
stable (thus it is not represented on the chart.)
11
Human Development Index

• These slides indicate the change in Human
  Development Index rankings from 2004 (using 2002
  data) to 2006 (using 2004 data). The Human
  Development Index measures achievements in terms
  of life expectancy, educational attainment and
  adjusted real income. Analysts divided countries
  into 3 rank groups-H-high human development Rank
  1-63 medium human development, 64-146 and low,
  147-177. UNDP, Human Development Report 2004,
  Cultural Liberty in Todays Diverse World,
  http//hdr.undp.org/reports/global/2004/?CFID8257
  709CFTOKENdfafd526399a4954-71537A04-1321-0B50-35
  D1567B0EAF4F44jsessionide630c4573eae406a5f65
  (177 countries)
• The 2006 index uses 2004 statistics. (177
  countries) . The 2007 report will be out in
  November 2007.

12
The Human Development Index rankings for EITI
countries have generally decreased. We expected
development to lag reforms and accountability.
13
But non-EITI countries showed mixed results (not
percentages) on development many regressed, but
some countries such as Ecuador and Egypt made
significant improvements.
14
Average Human Development Improvements (actual
results) The non-EITI average low development
made no improvement and thus is not represented
on the chart.
15
The Higher Level of Development, the Better
results for EITI and Non-EITI
16
Fiscal Policy Ratings

• These ratings are calculated by the IDA. If a
  country is not eligible for IDA assistance, we
  had no data on that country. Thus, this data set
  is less useful because of its short time frame
  and limited coverage. We hypothesized that EITI
  countries and non-EITI alike will have a hard
  time improving their governance, but EITI
  countries will focus on improving economic
  management, particularly related to fiscal policy
  and budget and financial management. This will
  allow them to do a better job monitoring and
  accounting for extractive industry revenues.

17
Fiscal Policy by Country-EITI. Only Cameroon and
Mauritania improved. EITI level one did not.
Most countries stayed stable or regressed.
18
Fiscal Policy-Non-EITI.Note that there is a lot
of missing data for non-EITI, as many of these
countries are not IDA recipients. Angola, Papua
New Guinea and Indonesia improved their
performance
19
Fiscal Policy Comparisons. Both EITI and non-EITI
improved their average ratings from a score of
3.35 to 3.5. However, when we compared medians,
the non-EITI did better. This may reflect better
governance, and less need for an external impetus
to promote good governance.
20
Quality of Budget and Financial Management

• These numbers are also based on the IDA ratings
  for 2005 and 2006. Here too if a country was not
  eligible for IDA funds, we had no data. Thus,
  this data set is less useful because of its short
  time frame and limited coverage.

21
Perceptions of Performance on Budget and
Financial Management, EITI. Ghana and Mauritania
improve, most countries are stable or regress.
22
It was hard to assess non-EITI countries because
of the limited data available. Of this group,
only Zambia improved its rankings.
23
Using means and medians, neither the EITI or
non-EITI improved their budget and financial
management performance in the 2 year period.
24
Voice and Accountability

• These ratings are devised by the World Bank and
  are available for 2000-2007 (except 2001).

25
This chart looks at EITI only, and compares
average and medians by EITI rankings.
26
As this and the next 2 slides show individual
country progressions by EITI grouping, it is as
important to note the ratings range as it is to
note the trend.
27
Some countries in rank 2 regressed, others
improved.
28
Some Rank 3 countries improved voice and
accountability dramatically with an end to war
and insurgency. (Sierra Leone and Liberia)
29
This is the percentage by which ratings have
changed over the 8-year sample period. For
example, if a country increased from 4.0 to 8.0
it would be reflected as the same percentage
change for a country going from 20.0 to 40.0.
Note that Azerbaijan and Nigeria, the supposedly
most advanced EITI countries, both experienced
decreases in average voice and accountability,
but DR Congo, Congo Republic, Liberia and Sierra
Leone had significant increases.
30
These are the averages and medians for EITI only
broken down by 2-year sample period. For
2000-2002 voice and accountability increases,
from 2002-2004 almost every voice and
accountability measure worsens, and from 2004 to
2006 progress is mixed, with medians doing better
than averages.
31
This chart compares voice and accountability
between Non-EITI averages and medians.
32
Non-EITI Voice and Accountability 2000-2006
33
Here are the changes in rating by percentile. It
is interesting to note that there has been much
less movement in the non-EITI countries overall.
This is perhaps because several EITI countries,
such as Sierra Leone and DR Congo, are emerging
from periods of conflict. In this group, conflict
continues in Sudan and Colombia, Russia, Egypt
and Indonesia have terrorist/insurgency threats
to stable governance.
34
Voice and accountability scores on average and by
median regressed for non-EITI Countries.
35
In this EITI/non-EITI comparison of voice and
accountability scores, the average and median
EITI improved more than non-EITI group of
countries.
36
Over the 2000-2007 period, EITI countries did
better than non-EITI on voice and accountability
medians and averages than non-EITI countries.
37
Corruption

• These ratings are also compiled by the World
  Bank, and use the same scale as the Voice and
  Accountability ratings.

38
Chart showing corruption scores for EITI
countries as a whole. Rank 2 countries appear to
be doing better on average than rank 1 EITI or
rank 3. This is probably due to level of
development and variance among countries rather
than any change arising from EITI. Also the
corruption scores rank perceptions, which may
lag actual improvements in corruption.
39
For rank 1, these numbers are low. We presume
this is because these countries
self-selected--they have great problems with
corruption and they may see EITI as an outside
panacea. Despite the sellers market for oil and
gas products, they may feel they must signal to
investors that they are trying to improve
governance.
40
In general, Rank 2 EITI countries seem to be
doing better on perceptions of corruption.
41
While some EITI rank 3 countries showed some
improvement in corruption perceptions, many
declined.
42
EITI performance on corruption indicators varied
greatly among countries. Azerbaijan, Cameroon,
and Kazakhstan made significant improvements.
With the end of conflict in Liberia, it made huge
improvements, its corruption rating improved by
1,260 in this period. Despite this huge
increase, the EITI group did not do well on
medians and on average.
43
This comparison of total change and change over
2-year periods among EITI countries does not show
corruption improvements as hypothesized. The
positive number in Group 1 is due to the small
sample size2 countries pulled up by Azerbaijan.
44
As this chart shows, non-EITI countries did
better on corruption.
45
Non-EITI corruption indicators 2000-2006.
46
Non-EITI corruption indicators by country.
47
As this chart shows, the averages and the medians
are improving for non-EITI, while they are
declining for EITI. We hypothesize this may be
because EITI countries are starting from a
position of weaker governance and greater
corruption and perception may lag improvements
or real improvements have simply not yet occurred
for the data reflecting perceptions of change.
48
EITI/non-EITI corruption comparison in 2 year
increments.
49
OECD Country Risk Ratings

• These ratings, measured quarterly (reported here
  are the January ratings for each year), are an
  indicator of investors confidence in the economy
  as a whole. They range from 0 (very high
  confidence) to 7 (very low confidence). Thus
  countries improve when the risk ratings decline.
  The classification of countries is achieved
  through the application of a methodology
  comprised of two basic components (1) the
  Country Risk Assessment Model (CRAM), which
  produces a quantitative assessment of country
  credit risk, based on three groups of risk
  indicators (the payment experience of the
  Participants, the financial situation and the
  economic situation) and (2) the qualitative
  assessment of the Model results, considered
  country-by-country to integrate political risk
  and/or other risk factors not taken (fully) into
  account by the Model. See
• http//www.oecd.org/document/49/0,2340,en_2649_341
  71_1901105_1_1_1_37467,00.html

50
While average EITI based on development levels
barely change, median country risk ratings do not
change at all, and are the worst they can be.
51
OECD Risk by EITI Group
52
OECD Risk medians by EITI Group
53
EITI Rank 1 Improving or Stable
54
Rank 2 Improving or Stable Kazakhstan is making
a significant improvement)
55
Most EITI countries receive country risk ratings
of 7 or 6, relatively high risk rankings.
Trinidad and Tobago is the only EITI country
ranked as Highly Developed under the UNDP Human
Development Index and that may explain why its
rating is so low (positive).
56
This chart shows that few EITI countries improved
their country risk rating over time.
57
On average, based on development, Non-EITI
countries performed slightly better on country
risk (stable or improving).
58
Non-EITI did better on country risk medians
(stable or improving)
59
Relatively more non-EITI countries have improved
their ratings over the 2000-2007 sample period,
making the average and median improvement higher
as well.
60
In general, non-EITI countries did slightly
better at improving their Country Risk ratings
than EITI countries, reflecting better individual
performance.
61
Non-EITI medians improve in 2004, while EITI
medians are stable.
62
OECD Risk Averages by EITI Grouping Compared to
Non-EITI
63
Medians compared to non-EITI medians non-EITI
does much better.
64
On average, EITI and non-EITI medium development
are doing better in improving their country risk
scores than the low development countries. This
probably reflects their governance expertise.
65
Low development countries have a hard time
improving their country risk scores.
66
Note that the median numbers really dont change,
except for the non-EITI median and the EITI rank
1 median, which has only 2 countries to sample
from. The EITI median rank 3 did not improve it
stayed at 7.
67
Food for thought (1)

• First sign of improvement-changes to business
  climate.
• Improvements depend on development levels, which
  correspond with governance ability and expertise
  as well as political will.
• One would think improvements to business climate
  would be reflected in similar improvements in
  fiscal policy/financial management, but we did
  not find that. This may reflect different
  measuring tools and objectives. In addition, the
  WB Doing Business Report ranks all countries,
  while the IDA data ranked only IDA eligible
  countries. The samples were not consistent.

• Every country is different, but its interesting
  to note we found increased voice and
  accountability in several EITI countries.
• Non-EITI did better than EITI in both lowering
  corruption and in improving country risk
  classifications EITI average on country risk
  remained stable, while non-EITI declined from a
  ranking of 6.2 to 5.8 during the 7 year period.
  Only 6 of the EITI group improved their rankings,
  with 2 countries having more than 1 point
  improvements. Meanwhile, 12 non-EITI improved
  their country risk rankings, with 3 countries
  making significant improvements.

68
Food for thought (2)

• What does this tell us about EITI? In a sellers
  market for oil (when countries such as China are
  perfectly willing to invest in countries such as
  Sudan despite political risk, corruption, and a
  lack of democratic accountable governance, why
  have 26 countries (1/2 of the extractives
  exporters) adopted EITI? They may be-
• Trying to signal to markets that they are really
  determined to change
• or they are so messed up they need outside forces
  and the EITI feedback loop to force change.
• Whatever the reason policymakers and citizens
  want to change perceptions of their country and
  want to change aspects of their governance
  behavior. The key lies in how effectively they
  will implement EITI and empower NGOs to monitor
  their efforts.
• This is why we encourage EITI to examine how it
  can work with the World Bank and individual
  donors to build a stronger demand within resource
  rich countries for transparency and
  accountability.