Вы находитесь на странице: 1из 1

A

TALK of the TOWN


PHILIPPINE DAILY INQUIRER

14

SUNDAY, JUNE 24, 2007


Please send feedback to
jsarmiento@inquirer.com.ph

Editor Juan V. Sarmiento Jr.

Party-list system: Mathematical absurdity


MAGINE A LAND WITHOUT MEANINGFUL ELECTIONS. THE
people, singing the songs of angry men, throw off a dictators
yoke, with nuns holding flowers and rosaries leading a peaceful
stand against tanks. Euphoric as democracy returns and basking in
global admiration, the people proclaim that they shall never again
be stripped of their sacred right of suffrage. In fact, they constitutionalize a radical new system that fills one-fifth of the House of
Representatives with marginalized sectors representatives.

Imagine, however, that when the fanfare


ends and the votes are tallied, over half the
seats in this special block of one-fifth are
left mysteriously empty. This allegory describes the Philippine party-list system,
and every unfilled seat represents a damning disenfranchisement and a dastardly
blow to the great spirit of social justice immortalized in the 1987 Constitution.
Our Constitutions
Article VI, Section 5
mandates that 20
percent of the House,
currently 55 seats,
shall be composed of
representatives
elected through a
party-list system of
registered national,
regional, and sectoral parties or organizations. After the
2004 elections, however, only 23 partylist representatives
were seated, or 10
percent of the House.
Reconciling conflicting visions in the
1986 Constitutional Commission (Concom), the Supreme Court in its 2001 Ang
Bagong Bayani decision ruled: The partylist system is a social justice tool designed
not only to give more law to the great
masses of our people who have less in life,
but also to enable them to become veritable lawmakers themselves.
Thus, the system must cater solely to national, regional and sectoral parties of the
marginalized and underrepresented. Ang
Bagong Bayani warned: [T]he law crafted
to address the peculiar disadvantages of
Payatas hovel dwellers cannot be appropriated by the mansion owners of Forbes Park.
This unifying constitutional philosophy,
however, does not cure the Party-List Acts
(Republic Act No. 7941) incomparably
vague implementing formula. Section 11
reads: [Parties] receiving at least two percent of the total votes cast for the party-list
system shall be entitled to one seat each:
Provided, That those garnering more than
two percent of the votes shall be entitled to
additional seats in proportion to their total
number of votes. In addition, no party
may receive more than three seats, a rule
that limits the House influence a party may
gain without fielding candidates in the regular district elections.

Veterans Federation Party decision that the


20-percent requirement was merely a maximum figure. This disregarded the Constitution, the supreme law, and upheld the
Party-List Act, an ordinary law.
Worse, the Veterans decision promulgated an electoral formula riddled with baseless mathematical errors that drive down
the total party-list seats allocated.

The Party-List Act has


a 2% threshold, but
55 seats x 2% is an
impossible 110%,
hence so many seats
are left unfilled

Impossible 2-percent threshold

The Commission on Elections soon discovered that it could not fill the required
20 percent of House seats. The culprit is
the Party-List Acts 2-percent threshold.
A party must receive at least 2 percent of
the votes to qualify for a seat. However, 55
seats multiplied by 2 percent is 110 percent
of the vote, clearly an impossible figure only Garci could meet.
The 2-percent threshold appears based
on the original 250 seats of the House20
percent of 250 is 50, and 100 percent divided by 50 seats is 2 percent per seat.
However, even 50 seats will be filled only in the idealized scenario in which exactly
50 parties each receive exactly 2 percent of
the vote. The moment a 51st party enters
and receives 0.01 percent of the vote (or if
one of the 50 parties receives 2.01 percent), the 50th seat will be left empty because the last 2 percent block cannot be
completed. The 2 percent thresholds
mathematical absurdity is increasingly obvious because the House expands to match
population growth.

German system

Some Concom delegates discussed the


German parliamentary electoral system,
which has a similar threshold designed to
ensure that winning parties have sufficiently large constituencies. This system,
however, allocates seats for the entire parliament and deals with a handful of the
countrys largest parties. This is completely
different from the Philippine party-list system, which deals with a large number of
very small parties.
Further, while the German system also
features a second vote at the national level,
the German formula connects results from
this second vote to the regular district elections, and assigns additional seats to parties
that receive a higher proportion of second
votes at the national level than first votes at
the district level, which empowers dispersed
constituencies. Lacking this crucial connection and given the completely different scale
of parties involved, the Philippine formula
has little in common with the German formula, contrary to the myth perpetuated in
existing legal scholarship.

Supreme Court confusion

Thus, the imported concept of a 2-percent threshold must be viewed with trepidation, especially given how the Supreme
Court has treated the three-seat cap, which
has no equivalent in the German system.
The Supreme Court soon found itself
caught between the Scylla of the Party-List
Acts 2-percent threshold and the Charybdis of the Constitutions 20-percent requirement. Appallingly, it ruled in its 2000

Errors

The Veterans formula begins by allocating seats to the party with the highest
vote. It allocates one
seat for each 2 percent
of the vote this first
party has obtained, up
to the three-seat cap.
This is its first error, as
Veterans fails to explain why the first party is subject to a separate formula, and fails
to even explain the rationale for this separate formula (which appears to be the mathematically absurd 2 percent per seat ratio).
The Veterans formula then allocates seats to
each other qualifying party. It assigns one seat
each then allocates additional seats by dividing
each partys vote by the first partys vote, then
multiplying the result by the first partys additional seats beyond its first (usually two). In its
second error and contrary to the Party-List Act
requirement, the Veterans formula is not proportional and does not form a rough straight
line when graphed. (See chart beside Figure 3)
(Using Buhays present 8.10 percent, the
formula would allocate no seats to parties
with 0-1.99 percent, one to those with
2.00-4.04 percent, and two to those with
4.05-8.09 percent.)
The Veterans formula allows only the first
party to receive the maximum number of
seats, its third error. Even if the second party
obtains just one vote less, it will still receive
one seat less. This breaks proportionality because such results are practically equal, since
one cannot allocate fractions of seats.
Worse, the strongest parties are irrationally
forced to compete because only one will be allotted the three-seat maximum no matter how
high their percentages.

Inconsistent

Further, the Veterans formula produces


inconsistent results that depend solely on
the first partys votes.
Its fourth error is that it continues to count
the first partys votes in excess of 6 percent, the
maximum considered by the separate first-party formula. That is, it allocates three seats to
the first party whether it has received 6 percent, 20 percent or 50 percent of the votes.
However, the higher the first partys
votes, the less seats the Veterans formula
allocates to all other parties, and we are
only beginning to observe the catastrophic
results now that very strong parties with
disproportionately high percentages are
emerging. For example, if the first party obtains 20 percent of the votes, other parties
will be allocated two seats only if they obtain at least 10 percent.
Finally, as the Veterans formulas fifth error,
its two subformulas are inconsistent. If the first
party obtains exactly 6 percent, other parties
will receive two seats if they obtain at least 3
percent, which contradicts the first-party formula (which requires 4 percent for two seats).
At present, Buhay leads with 8.10 percent and
the Veterans formula would allocate only two
seats to second placer Bayan Muna despite its
very high 6.59 percent, or enough for three
seats under the separate first-party formula.
Further, were Buhays percentage lower,
say 6 percent, the inconsistent formula
would allocate two seats instead of one to
Apec, with 3.50 percent.

Proposed formula

A proper party-list formula need only


fulfill two simple goals: 1) It must fill up 20
percent of House seats; and 2) Seat allocation must be proportional, including allocations to losing parties.
In this proposed formula, one selects a
party, then divides each partys votes by the
chosen partys votes. One then drops decimals from the resulting ratios, and assigns
that many seats to each party, up to the
three-seat maximum. If the total number of
seats assigned exceeds 55, one chooses another party with less votes, and vice-versa.
If no divisor yields exactly 55 seats, one
assigns the empty seats to parties with the
highest decimal components in their ratios,
ignoring the seats already allocated. This
remains proportional because excess
votes represented in each partys decimal
component still give that party an equal
chance to obtain an additional seat. (This
tiebreaker function is similar to the German systems Niemeyer tiebreaker, which
cannot be used whenever the 2-percent
threshold is used, as the number of empty
seats is so high that there are simply no ties
to break and the result becomes absurd.)
This proposed formula creates an implicit threshold based on how parties votes
are distributed. Should one wish to add a
fixed threshold (say, 100,000 votes), one

Figure 1: Separate Veterans first-party formula


Percentage of votes
0 - 1.99
2.00 - 3.99
4.00 - 5.99
6.00 or higher

Problem: Veterans does not explain why there is a separate formula for the first party. It also does not explain the
rationale behind this separate formula.

Seats
0
1+ 0
1+ 1
1+ 2

(Note: The notation 1+0, 1+1 and 1+2 instead of 1, 2 and


3 is intentional because the Veterans formula returns additional
seats after the first. This actually creates a mathematical problem
not discussed in the article.)

Figure 2: Veterans formula seat allocation for 2007 elections


Problems:
Only the first party can get the maximum three seats, because the
Veterans formula's result can never reach two additional seats for
any other party. (The ratio on the left is always less than one.)
The Veterans formula continues to count the first party's votes in
excess of the maximum 6% counted by the separate formula for the
first party. This drives down every other party's seat allocation.
The first party formula and the formula applied to other parties are
inconsistent.
The Veterans seat allocation (left) is not proportional, especially
when one factors the allocation to losing parties. Consider the
example where the first party obtains exactly 6% of the vote. (See
Figure 3)

Percentage
of vote of party
8.10% (Buhay
percentage)

2 seats (additional
seats allocated
to Buhay)

Figure 3: Veterans seat allocation if first party obtains 6% (not proportional)


Percentage of votes
0 - 1.99

Seats
0

3.00 - 5.99

1+ 1

3.5
3

2.5

1+ 0

2.00 - 2.99

Seats

By Oscar Franklin B. Tan

1+ 2 (first party only)

6.00

1.5
1

0.5
0

4
Votes

Figure 4: Veterans formula seat allocation given Buhay's 8.10% (not proportional)
Percentage of votes

Seats

0 - 1.99

2.00 - 4.04

1+ 0

4.05 - 8.09

1+ 1

8.10

1+ 2 (Buhay only)

Problem: The Veterans formula is inconsistent and drives


down other parties' seat allocations using Buhay's high
percentage. Bayan Muna, with 6.59%, would have obtained three seats under the separate first-party formula.
Apec, with 3.50%, would have obtained two seats if
Buhays percentage were lower.

Figure 5: Proposed seat allocation (using Agham's 1.03% as benchmark)


Percentage of votes

Seats

0 - 1.02

Partys percentage

1.03 - 2.05

2.06 - 3.08

3.09 or higher

1.03% (Aghams
percentage)

Seat allocation
(drop fractions,
subject to
tiebreaker
function)

Figure 6: Application of proposed formula using latest Comelec data (as of June 9)
TOTAL

13,719,165

Party
Buhay
Bayan Muna
Cibac
Gabriela
Apec
A Teacher
Akbayan
Butil
Alagad
Batas
Coop-Natco
Anakpawis
Abono
Agap
ARC
An Waray
FPJPM
Amin
ABS
Kabataan
Aba-Ako
Senior Citizens
Kakusa
VFP
Uni-Mad
Anad
Banat
Abakada
Bantay
1-Utak
Cocofed
Agham

Party's
votes
1,111,035
903,600
720,153
566,571
479,608
446,929
420,910
403,209
402,918
360,277
341,971
340,392
333,603
319,311
295,499
270,791
258,773
252,250
213,927
211,074
205,344
198,948
190,368
183,779
182,354
169,525
167,222
161,141
160,568
154,589
145,176
141,773

55

20

Party %
of vote

Proposed
formula
ratio

Proposed
formula
seat assign

Supreme
Court
formula ratio

Supreme
Court
seat assign

8.10
6.59
5.25
4.13
3.50
3.26
3.07
2.94
2.94
2.63
2.49
2.48
2.43
2.33
2.15
1.97
1.89
1.84
1.56
1.54
1.50
1.45
1.39
1.34
1.33
1.24
1.22
1.17
1.17
1.13
1.06
1.03

7.84
6.37
5.08
4.00
3.38
3.15
2.97
2.84
2.84
2.54
2.41
2.40
2.35
2.25
2.08
1.91
1.83
1.78
1.51
1.49
1.45
1.40
1.34
1.30
1.29
1.20
1.18
1.14
1.13
1.09
1.02
1.00

3 (7)
3 (6)
3 (5)
3 (4)
3
3
2+1
2
2
2
2
2
2
2
2
1+1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

2.00
1.63
1.30
1.02
0.86
0.80
0.76
0.73
0.73
0.65
0.62
0.61
0.60
0.57
0.53
0.49
0.47
0.45
0.39
0.38
0.37
0.36
0.34
0.33
0.33
0.31
0.30
0.29
0.29
0.28
0.26
0.26

1+ 2
1+ 1
1+1
1+1
1+0
1+0
1+0
1+0
1+0
1+0
1+0
1+0
1+0
1+0
1+0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

simply does not assign seats to parties below this threshold. Similarly, the three-seat
cap may be increased and the formula
works the same way.

Solving the problem

Applying the formula, one chooses Agham


and divides all parties percentages by its 1.03
percent. This yields 53 seats, plus two assigned to Akbayan and An Waray because of
their very high decimal components of 0.97
and 0.91, respectively. (Dividing all parties
votes by 135,000 yields the same result.)
This result is proportional because each
party receives one seat for each 1.03 percent,
and the formula ignores the strongest parties
votes beyond the three-seat cap so as not to
drive down other parties allocations.
The Veterans formula, in contrast, will allocate only a piddling 20 seats, with only Buhay

receiving the full three seats. Note that although


the proposed formula assigns seats to parties
with only 1 percent, this is because it factors the
vote dispersal unique to the Philippine system
and its large number of small parties.
It is imperative that the mathematically
absurd 2-percent threshold be struck
down, whether by the Supreme Court or by
legislators, and the Veterans formula abandoned given how its flaws are all the more
evident with a first party such as Buhay obtaining far more votes than the others.
This threshold aside, it is clearly possible
to allocate the full 20 percent of the House
promised by the Constitution in a proportional manner. We bewail bogus party-list
groups and confused listings of parties at the
precincts, but we must also reform the electoral formula itself to enable party-list representatives to actually sit in Congress. Our

Solution: If the
mathematically absurd 2% threshold
is struck down, the
proposed formula
completely fills up
the constitutionally
required 20% of the
House in a proportional manner, factoring in the vote
dispersion unique
to the Philippine
party-list system.
The strong parties
high votes do not
drive down other
parties votes.

92

Number
of party-list groups

13.7
million

Count of votes
for party-list groups
as of June 9

Constitution must be protected from a dictators lust for power and our laws mathematical shortcomings.
(Oscar Franklin B. Tan graduated from Harvard Law School on June 7, where he was chosen to speak at the commencement ceremony,
and from the UP College of Law in 2005, where
he chaired the Philippine Law Journal. He
graduated from the Ateneo de Manila in 2001,
majoring in Management Engineering and
Economics. This article is adapted from The
Philippine Party-List Experiment: A Tragedy of
Flawed Mathematics and Policy (Philippine
Law Journal, Volume 78, page 735), his 76page-freshman-year article informally supervised by Dean Pacifico Agabin and awarded UP
Laws first Justice Vicente V. Mendoza legal
writing prize. The article can be downloaded at
http://www.xavier97.com/oscar.)