This action might not be possible to undo. Are you sure you want to continue?
Party-list system: Mathematical absurdity
By Oscar Franklin B. Tan
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 Constitution’s 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 Act’s (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.
TALK of the TOWN
PHILIPPINE DAILY INQUIRER
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 Seats 0 1+ 0 1+ 1 1+ 2
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 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 party’s vote by the first party’s vote, then multiplying the result by the first party’s 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 Buhay’s 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.
SUNDAY, JUNE 24, 2007
Please send feedback to email@example.com
Editor Juan V. Sarmiento Jr.
MAGINE A LAND WITHOUT MEANINGFUL ELECTIONS. THE people, “singing the songs of angry men,” throw off a dictator’s 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.
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.
(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) X 2 seats (additional seats allocated to Buhay)
• • •
The Party-List Act has a 2% threshold, but 55 seats x 2% is an impossible 110%, hence so many seats are left unfilled
Figure 3: Veterans seat allocation if first party obtains 6% (not proportional)
Percentage of votes 0 - 1.99 2.00 - 2.99 3.00 - 5.99 6.00 Seats 0 1+ 0 1+ 1
Seats 3.5 2.5 1.5 0.5 0 2 3 4 Votes 5 6 1 2 3
1+ 2 (first party only)
Figure 4: Veterans formula seat allocation given Buhay's 8.10% (not proportional)
Percentage of votes 0 - 1.99 2.00 - 4.04 4.05 - 8.09 8.10 Seats 1+ 0 1+ 2 (Buhay only) 1+ 1 0 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 Buhay’s percentage were lower.
Figure 5: Proposed seat allocation (using Agham's 1.03% as benchmark)
Percentage of votes 0 - 1.02 1.03 - 2.05 2.06 - 3.08 3.09 or higher Seats 0 1 2 3 Party’s percentage 1.03% (Agham’s percentage) X Seat allocation (drop fractions, subject to tiebreaker function)
The Commission on Elections soon discovered that it could not fill the required 20 percent of House seats. The culprit is the Party-List Act’s 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 House—20 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 threshold’s mathematical absurdity is increasingly obvious because the House expands to match population growth. 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 country’s 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. 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 Act’s 2-percent threshold and the Charybdis of the Constitution’s 20-percent requirement. Appallingly, it ruled in its 2000
Impossible 2-percent threshold
Supreme Court confusion
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 party’s votes by the chosen party’s 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 party’s decimal component still give that party an equal chance to obtain an additional seat. (This tiebreaker function is similar to the German system’s 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
Further, the Veterans formula produces inconsistent results that depend solely on the first party’s votes. Its fourth error is that it continues to count the first party’s 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 party’s 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 formula’s 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 Buhay’s percentage lower, say 6 percent, the inconsistent formula would allocate two seats instead of one to Apec, with 3.50 percent.
Figure 6: Application of proposed formula using latest Comelec data (as of June 9)
TOTAL 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 13,719,165 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 Party % of vote 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 Proposed formula ratio 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 55 Proposed formula seat assign 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 Supreme Court formula ratio 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 20 Supreme Court seat assign 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
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.
Number of party-list groups
Count of votes for party-list groups as of June 9
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
Solving the problem
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.
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
Constitution must be protected from a dictator’s 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 Law’s first Justice Vicente V Mendoza legal . writing prize. The article can be downloaded at http://www.xavier97.com/oscar.)
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.