Myanmar: A Pre-Election Primer

Prepared for the Conflict Prevention and Peace Forum by Richard Horsey1 (18 October 2010)

I. Introduction
With the Myanmar election only three weeks away, this paper aims to give a concise
overview of the state of play: the constituencies, political parties and candidates. It also
attempts to predict the range of possible outcomes of the ballot.
Any such prediction is extremely difficult, but it is possible to sketch certain plausible
scenarios, and rule out certain others. Thus, it is clear that a repeat of the 1990 democratic
landslide is statistically impossible. It is also difficult to say whether there will be
systematic irregularities on election day that would influence the outcome. However, while
there will undoubtedly be some irregularities, a fraudulent vote count is on balance unlikely
– both because of the decentralized and semi-public nature of the count, and because the
odds have already been stacked in favour of the military/USDP.
The analysis provided in section IV suggests that there can be no “democratic”
majority in either house of the national legislature, but that, unless there is major ballot
fraud on election day, the military/USDP bloc is also unlikely to have a majority in either
house. The most likely outcome is that the National Unity Party (NUP) will hold the
balance of power between the two dispensations. This would give the NUP a powerful role,
and it is unclear how it will use its legislative influence. The party is undoubtedly part of
the political ‘establishment’, but its policies are unlikely to coincide with those of the
present regime on all issues – it would not be surprising if it decided to push for one of the
presidential nominees to be drawn from its ranks, for example. If the NUP does end up
holding the balance of power in the legislatures, this may mean that lawmaking is
dominated by a conservative, authoritarian-leaning nationalism; but it would certainly not
be merely a facsimile of the present regime in civilian clothing.
II. Constituencies
The Election Commission designated a total of 1163 constituencies for this electoral cycle,2
as follows:
Pyithu Hluttaw (Lower House): 330 constituencies
Amyotha Hluttaw (Upper House): 168 constituencies
14 State/Region legislatures: 665 constituencies (636 township constituencies,3
plus 29 special ethnic constituencies)
1 Richard Horsey is an independent political analyst and a former ILO liaison officer in Myanmar; he is fluent in
the Burmese language.
2 For a detailed analysis of constituencies, see Richard Horsey, “Countdown to the Myanmar Elections”, Conflict
Prevention and Peace Forum Briefing Paper, 24 August 2010.
2
On 16 September 2010, the Election Commission announced that elections would not
take place in certain ethnic areas of the country “as they are in no position to hold free and
fair elections”. Most of these places are insecure or conflict areas.
According to these announcements, no voting will take place in four whole townships
(controlled by the United Wa State Army ceasefire group) and some 300 village-tracts.
These areas are spread across thirty-two townships. This means that only six constituencies
have been affected in toto: the four lower house constituencies corresponding to the excluded
Wa townships (no State/Region constituencies were designated for these townships), and
two State/Region constituencies in Kachin State (Ingyangyang 2 and Sumprabum 2).
Therefore, on 7 November elections will take place in 1157 constituencies, as follows:
Pyithu Hluttaw (Lower House): 326 constituencies
Amyotha Hluttaw (Upper House): 168 constituencies
14 State/Region legislatures: 663 constituencies
In terms of population, the 300 excluded village tracts have at most 190,000 eligible
voters.4 The four excluded Wa townships have at most an additional 150,000.5 Together,
this represents about one percent of all eligible voters in the country. In terms of ethnic
exclusion, the refusal of the Election Commission to register the three Kachin parties, and
its subsequent rejection of independent candidates linked to these parties, is a far more
serious issue (in the case of the Wa, it was not so much that the Election Commission
excluded them; rather, the United Wa State Army indicated that, due to its rejection of the
2008 constitution, and in particular the provisions for the Wa region, it would not allow
elections to take place in the areas under its military control).
In 1990, the areas that were excluded from the ballot for security reasons were much
more extensive. At that time, elections did not take place in seven out of 492 constituencies.
However, much larger geographical areas (and hence populations) could not participate due
to security reasons, since at that time, significant swathes of the borderlands were under
the territorial control of armed opposition groups, and were therefore never designated as
separate constituencies.
3 That is, two constituencies per township, except that no State/Region constituencies were designated for the
eight townships of Naypyitaw (which are ‘Union territory’ under the direct administration of the president) or
for the four townships in Shan State under the control of the United Wa State Army ceasefire group (which may
also be declared ‘Union territory’ on grounds of security). There are 330 townships in Myanmar, so the total
State/Region constituencies is (330 – 8 – 4) = 318 x 2 = 636 constituencies. This is the number listed in Union
Election Commission Notification No. 88/2010 (11 August 2010).
4 The total number of village tracts/wards in the country is 16,000, so 300 represent less than two percent. These
are mostly remote, mountainous, conflict affected parts of the country, with a population density far below the
national average. Assuming the population density is one-third of the average, the number of eligible voters
would be ((300/16,000) x 30,000,000)/3 = 188,000.
5 The population of the Wa special region (six townships) is about 450,000 (UNODC, 2006). Assuming even
population density, the four excluded townships would have a population of 300,000. Assuming eligible voters
make up fifty percent of the population (the national average), that would give 150,000 voters.
3
III. Political parties and candidates
A total of thirty-seven political parties will compete in the elections (see list in appendix 1).
Ten of the forty-seven parties who applied to the Election Commission for registration6 were
not approved:
1. Kachin State Progressive Party Not approved by Election Commission
2. Northern Shan State Progressive Party Not approved by Election Commission
3. United Democracy Party (Kachin State) Not approved by Election Commission
4. People’s New Society Party Not approved by Election Commission
5. All National Races Unity and
Development Party (Kayah State)
Failed to complete registration process
6. Mro National Party constituencies Failed to complete registration process
7. Myanmar Democracy Congress Failed to contest minimum 3 constituencies
8. Myanmar New Society Democratic Party Failed to contest minimum 3 constituencies
9. Regional Development Party (Pyay) Failed to contest minimum 3 constituencies
10. Union Kayin League Failed to contest minimum 3 constituencies
Of the thirty-seven political parties that will compete, most are small regional parties
which have nominated few candidates. The list in appendix 1 gives details on how many
approved candidates each party will field. These numbers are based on information
provided by the parties themselves, and are incomplete, since the Election Commission has
never published a consolidated list of approved candidates for the country as a whole.
It can be seen that most parties have fielded relatively few candidates. The Union
Solidarity and Development Party and the National Unity Party are each contesting almost
all constituencies, and each have more candidates that all other parties combined. This
means that there will be very few constituencies where the USDP runs unopposed.
However, in many constituencies – particularly rural regional constituencies in central
Myanmar – the choice will be between two candidates representing the party linked with
the current regime (the USDP) and the party linked with the pre-1988 socialist regime (the
NUP).
In the absence of an official national list of the candidates running in each
constituency, it is impossible to give a definitive analysis. However, an unofficial list of
candidates for the upper, lower and local seats in Yangon Region is available, and can give
some idea of how things are playing out in the most important political region.
6 See list in appendix 1 of “Countdown to the Myanmar Elections”, Conflict Prevention and Peace Forum
Briefing Paper, 24 August 2010.
Yangon Region candidate distribution:
Total seats: 149 (12 Upper House, 45 Lower House, 92 Region legislature)
Total candidates: 454 (48 Upper House, 195 Lower House, 211 Region legislature)
Parties contesting: 14
Independent candidates: 26
Seats with only one candidate: 1 (no ballot; candidate wins by default)
Seats with only USDP and NUP candidates: 63 (only in Region legislature)
4
This is the part of the country where political competition is most intense. Even here,
though, voters will have only a choice of a USDP or an NUP candidate in two-thirds of the
seats in the Region legislature. For upper and lower house seats, however, there will be
considerably more choice, with a candidate from at least one of the main “democratic”
parties (the Democratic Party and the National Democratic Force) standing in nearly all of
these constituencies. A full breakdown is given in appendix 2.
By the numbers: comparison with 1990
2010 1990
Number of constituencies:7 1163 492
Number of constituencies in which elections held: 1157 485
Number of parties applied: 47 235
Number of parties contesting: 37 93
Total number of candidates: c. 3153 2296
Average number of candidates per constituency: 2.6 4.7
Total number of eligible voters: c. 30,000,000 20,800,000
Voter turnout (incl. invalid votes) ? 72.6%
IV. Possible outcomes
It is clear that in the run-up to the elections, the playing field has been tilted strongly
in favour of the USDP. Beyond this, it is impossible to know how much manipulation of the
vote will occur on election day. It is very likely that there will be individual instances of
intimidation, manipulation and fraud. But it is quite conceivable that there will be no
systematic attempts to rig the count in favour of the USDP, for two main reasons. First, the
numbers are already stacked heavily in favour of the regime: twenty-five percent of
legislative seats are reserved for military appointees, the USDP is contesting almost all
constituencies, and the main “democratic” parties are contesting less than half. This means
that a repeat of the embarrassing 1990 democratic landslide is statistically impossible.
Second, the election laws provide for counting of the votes at each individual polling station,
in the presence of candidates or their representatives and members of the public. This
makes it rather difficult to carry out a major manipulation of the count of the kind
witnessed in the 2008 referendum. The most obvious opportunity for manipulation is with
absentee voting, but it remains to be seen whether the number of such votes will be large
enough to significantly alter the outcome of the election as a whole; prima facie, it seems
unlikely.
In the absence of opinion polls or other reliable ways to quantify voter sentiment, it is
impossible to attempt any detailed predictions of the election outcome. However, it is
possible, on the basis of certain reasonable assumptions, to sketch two scenarios – a ‘best’
and ‘worst’ case, from the perspective of the democratic parties and the establishment
parties. The numbers in what follows are rough approximations. Although the Election
Commission has released no official candidate lists, it is known that the USDP and NUP
7 There are 1171 constitutionally-mandated constituencies. However, the Election Commission did not designate
eight Shan State Hluttaw constituencies for four townships in the Wa area, giving a total of 1163 for this
election cycle.
5
will compete in almost all constituencies, and most of the other parties have made the lists
of their own candidates available. Taken together, this information can allow some general
conclusions to be drawn.
Scenario 1: “Democratic maximalist”
This scenario assumes that the USDP and NUP will be systematically punished by
voters: in any constituency where there is a third choice, that candidate will win. In those
constituencies where there are only USDP and NUP candidates, each party is assumed to
have an equal chance of winning, with the choice coming down to the reputation of the
individual candidate.
On this scenario, the approximate division of national legislative seats will be as follows:
Upper House (Amyotha Hluttaw): Military 25%; USDP 14%; NUP 11%; other parties
50%
Lower House (Pyithu Hluttaw): Military 25%; USDP 19%; NUP 18%; other parties
38%
This could give ‘other parties’ a wafer-thin majority in the upper house, assuming that
they won every seat where a third-party candidate was contesting (unlikely) and assuming
that all such candidates would vote as a bloc (also unlikely). Note, however, that the
military and USDP together also fail to have a majority in the upper house. This implies
that on this scenario, the NUP would be likely to hold the balance of power.
In the lower house, it is statistically impossible for the ‘other parties’ to obtain a
majority. On the assumption that the USDP and NUP equally share the seats where only
they are contesting, and win no other seats, the military and USDP together will also fail to
reach a majority. Again, the NUP would hold the balance of power.
On the important question of choosing presidential nominees and electing the
president, the situation is slightly different. Recall that according to the constitution, three
nominees are selected – one by the elected representatives of the upper house, one by the
elected representatives of the lower house, and one by the military appointees of both
houses. The ‘other parties’ would therefore require a simple majority of elected seats in each
house (that is, not including military appointees) in order to choose the presidential
nominee of that house. On the present scenario, the ‘other parties’ would have two-thirds of
the elected seats in the upper house, but less than half of the elected seats in the lower
house. Therefore, they have a good chance of being able to choose the upper house
presidential nominee, but not the lower house nominee, for which the NUP will hold the
balance of power.
As for the electoral college itself, which will elect the president from among the three
nominees, what is important is the total proportion of seats held by each bloc in the
bicameral legislature as a whole. On the present scenario, the breakdown would be:
Presidential electoral college: Military 25%; USDP 17%; NUP 16%; other parties 42%
The NUP would therefore hold the balance of power in electing the president.
6
Scenario 2: “Equal shares”
This scenario assumes that in each constituency there is an equal chance of the USDP,
NUP and ‘other party’ candidates being elected. This is probably the best that the USDP
and NUP can hope for, short of vote rigging. The approximate division of national legislative
seats would be as follows:
Upper House (Amyotha Hluttaw): Military 25%; USDP 30%; NUP 27%; other parties
18%
Lower House (Pyithu Hluttaw): Military 25%; USDP 33%; NUP 30%; other parties
12%

This would give the military and USDP together a small majority in both houses.
However, they would not have a majority of the elected seats in either house, meaning that
for the selection of presidential nominees, the NUP would hold the balance of power in both
houses. In terms of the presidential electoral college, the breakdown would be:
Presidential electoral college: Military 25%; USDP 32%; NUP 29%; other parties 14%
On this scenario, the military and the USDP together would have a comfortable majority for
electing the president.
Both of these scenarios are extremes, and the outcome is likely to be somewhere in
between the two. The following general conclusions can be drawn.
1. The ‘other’ parties will not have a majority in either house. They can expect 18 – 50
percent in the upper house, and 12 – 35 percent in the lower house.
2. These figures probably overstate the size of any democratic bloc, since they include a
diverse range of democratic, ethnic and other parties (and a few independent
candidates) who may not vote together and may include (a few) parties or candidates
whose sympathies are closer to the ‘establishment’ parties than to the ‘democratic’
parties, or who might be co-opted.
3. Unless the vote-count is rigged, it is unlikely that the military and USDP together
will have a majority in either house. It is likely that the NUP will hold the balance of
power in both houses.
4. The ‘other’ parties have a reasonable chance of being able to select one of the three
presidential nominees (the upper house nominee). It is unlikely that the USDP will
have enough seats to be able to choose the lower house nominee. It is very likely that
the NUP will hold the balance of power in selecting the lower house presidential
nominee, and possibly the upper house nominee as well. (Under the constitution, the
third nominee will be chosen by the military bloc.)
5. The NUP may also hold the balance of power in electing the president from among
the three nominees. However, there is a reasonable chance that the USDP will
secure enough seats so that it, together with the military, will be able to make that
choice.
7
V. Conclusions
It is impossible to accurately predict how voters will vote on 7 November. Any analysis is
also frustrated by the fact that the Election Commission has not published consolidated lists
of approved candidates for the country as a whole. However, the distribution of candidates
can be determined to a reasonable approximation due to the fact that the two largest parties
(the USDP and NUP) will contest nearly all constituencies, and most of the other parties
have made their candidate lists available.8
Given that in a large number of
constituencies, only two candidates are
standing – one from the USDP and one from
the NUP – it is possible to make some general
predictions about the result of the elections in
the national legislatures.
The most important point to come out of
the analysis is that the NUP will probably hold
the balance of power between the
military/USDP and the other parties, in
multiple ways: in the lower house, and possibly
the upper house as well, and likely in several
regional legislatures. The NUP will also hold
the balance of power in selecting one or
perhaps two of the three presidential
nominees. And it may even hold the balance of
power in electing the president (although there
is a good chance that the military/USDP will
have enough seats to determine this).
The standard assumption in most
commentaries and analyses of the elections has
been that the NUP is just another proxy for the
regime, like the USDP. If that is the case, then
the present conclusions are of little interest,
since they imply that the military and its USDP/NUP proxies will easily control most of the
important levers of power after the elections. But the pivotal position that the NUP might
occupy should at least prompt some speculation about what would happen if the standard
assumption were wrong.
In fact, while the NUP is certainly not a natural ally of the National Democratic
Force and other democratic parties, there are reasons to believe that it is much more
independent of the regime and the military than is commonly assumed. In particular, on
social and economic issues its policies could diverge significantly from those of the
military/USDP. And if the party does end up holding the balance of power, it should come as
no surprise if it were to demand significant concessions – in terms of policy and influence –
in return for its support to the military/USDP bloc.
8 For profiles of all the political parties, see Richard Horsey, “Overview of registered political parties in
Myanmar”, Conflict Prevention and Peace Forum Briefing Paper, 15 June 2010.

Ref: rh2 << Download pdf

http://www.networkmyanmar.org/images/stories/PDF5/rh2.pdf

kai

About kai

Kai has written 913 post in this Website..

Editor - The Myanmar Gazette || First Amendment – Religion and Expression - Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.