PAS recently issued an open invitation to Anwar Ibrahim to contest the Kota Baru parliamentary seat in Kelantan.

While this may seem attractive to some quarters, especially those within Parti Keadilan Rakyat (PKR) who like the idea of a 'safer' seat for Anwar, I want to argue that this strategy, while probably beneficial to him personally, would be detrimental to the party's public image, its long term direction and most importantly, the future of opposition politics in Malaysia.

What differentiates PKR from PAS and DAP is that it is a Malay-led political party which is genuinely multi-ethnic. PAS, while theoretically multi-ethnic (one of its state exco members in Kelantan in Anwar Tan, a Chinese convert to Islam), is not open to non-Malays who are not Muslims.

DAP, while also theoretically multi-ethnic, is largely led by Chinese leaders with some Indian representation (though to be fair to the DAP, it is the only opposition party to have Malays, Chinese and Indian representatives at state and Parliament level).

PKR started out as Parti Keadilan in 1999 at a time when the rakyat was hungry for a new brand of politics, a major component of which was its ability to cross the racial divide and present a 'race-blind' approach towards combating the problems of 'cronyism, corruption and nepotism'. While the momentum that lead to the creation of the party has died down, I believe that the founding ideals have not.

Furthermore, the political 'space' in opposition politics is such that PKR is occupying a strategic and important 'middle ground' (or cynics might view it as the party having nowhere else to go). This 'middle ground' of taking a racially moderate or inclusive approach naturally translates to the 'middle ground' of mixed seats in electoral contests.

It is no surprise that PKR is 'left' with this middle ground of mixed seats. A typical parliamentary seat that it contested in the 2004 general election in Peninsular Malaysia would have had a composition of 58 percent Malay, 31 percent Chinese and 11 percent Indian voters. Compare this to75 percent Malay, 18 percent Chinese, 7 percent Indian for PAS-contested seats; and 27 percent Malay, 62 percent Chinese, 11percent Indian for the DAP.

The 'mixed' seats are also typically where the Barisan Nasional (BN) demonstrates its greatest electoral strength, hence the reason why PKR is 'left' to compete in these seats.

In 2004, in parliamentary seats where no ethnic group comprised more than 60 percent of the electorate, BN's share of the popular vote was 70 percent. In seats where Malays comprised of more than 80 percent of the electorate, BN's share of the popular vote was 57.6 percent. In seats where Chinese comprised more than 70 percent of the electorate, BN's share of the popular vote was only 46.3 percent.

Mixed seats have become more important in electoral configuration especially since the number and percentage of mixed seats increased after the 2003 electoral delimitation exercise.

'Coat-tail' effect

Hence, if PKR is left to occupy the middle ground of Malaysian politics and if it is to make any dent in BN's formidable electoral armour, its most prominent and popular leader has to contest in a mixed seat. If PKR's main message is that it wants to represent a new brand of politics which can cross the ethnic divide and can attract votes from all groups, would it not be hypocritical for its leader to contest in a seat like Kota Baru which is 79 percent Malay?

Furthermore, by choosing to contest in Kelantan, Anwar would be, directly or indirectly, signalling his desire to help PAS hold to on the state rather than trying to build the foundations of a strong opposition presence in multi-ethnic, urban areas.

If there is anyone in the opposition who can even attempt to make some headway in mixed seats, it would be Anwar. One has to remember that at the height of its national popularity or prominence in 1999, all of the party's parliamentary seats were won in seats where the Malay electorate was more than 60 percent. At 67 percent Malay, Permatang Pauh was the most 'mixed' of the party's five parliamentary seats in 1999.

All of its more prominent national leaders failed to win any seat in the mainly urban and mixed seats where they contested although Chandra Muzzafar, Zainur Zakaria and Ezam Noor came close in Bandar Tun Razak, Lembah Pantai and Shah Alam respectively.

None of these candidates can boast the same kind of national popularity, voter mobilisation potential, oratorical skills and political experience which Anwar can bring to the table.

He is arguably the only candidate in PKR, if not within the entire opposition, who can command a credible amount of Malay and non-Malay votes in whichever mixed seat he chooses to contest. His reputation among non-Malay voters would have increased especially given his recent stand against the National Economic Policy.

Anwar could command a significant minority of Malay voters just because he is Anwar. While it is still far from guaranteed that he can win in a mixed seat in an urban area just by showing up (hence the argument to search for a safer seat for him), he still presents the best candidate for PKR to win a mixed seat.

Furthermore, one cannot ignore the potential 'coat-tail' effect which an Anwar candidacy might have in the seats immediately adjoining the one that he chooses to contest. While not an exact comparison, one cannot help but point to the Lim Kit Siang effect in the Kinta Valley seats of Ipoh Timor, Ipoh Barat and Batu Gajah in 2004 even while the national sentiment swung firmly in favour of Abdullah Ahmad Badawi and the BN.

Would not Anwar, given his prominence and stature, have an even bigger effect in the next elections, especially if he is surrounded by other strong PKR (and perhaps DAP and PAS) candidates in nearby parliamentary and perhaps state seats?

Where then?

After putting forth some arguments with regard to 'why' Anwar shouldn't contest in Kota Baru but rather in a mixed urban seat elsewhere, the question then naturally turns to 'where'.

Again, following the argument that PKR should occupy the middle ground of opposition politics and hence mixed seats, it is only natural that its leader should target states which are also mixed in its ethnic composition. This would exclude predominantly Malay states like Perlis, Kelantan and Terengganu.

It would probably exclude Kedah (75 percent Malay) and Pahang (almost 70 percent Malay). Assuming that he would choose to contest in Peninsular Malaysia rather than Sabah (another state that PKR is targeting), this leaves Penang, Perak, Selangor, Federal Territory of Kuala Lumpur (KL), Negri Sembilan, Malacca and Johor.

One can immediately dismiss Johor as an option given BN's overwhelming dominance in this state. (There's a difference between committing political suicide and taking a calculated risk.) One could probably exclude Negri Sembilan and Malacca given that these are relatively small states and the potential coat-tail effect of an Anwar candidacy would be limited.

While Perak may at first sight seem to fit the bill - it is ethnically mixed and it has the largest number of parliamentary seats in Peninsular Malaysia after Johor - after a more careful examination, one quickly realises that there are no mixed seats which can be classified as urban seats and has a potential of voting opposition.

Taiping and Beruas are likely to be contested by the DAP which has a more established presence in these areas, Gopeng is highly unwinnable and its semi-urban nature means that the kinds of dissatisfaction found in other urban areas (such as urban inflation, issues of corruption and so on) would be less salient here.

Parit Buntar, won by PAS in 1999, is 67 percent Malay and is largely a rural area. Kuala Kangsar, also 67 percent Malay, is currently occupied by cabinet minister Rafidah Aziz ( photo ), who brings along all of the incumbency advantages which comes with occupying that position.

The discussion of seats in Perak reveals a more subtle point. Firstly, even though I have strongly recommended that Anwar contests in a mixed seat (where no one ethnic group comprises more than 60 percent of the electorate), political reality dictates that Anwar cannot contest in a 'mixed' seat with a Chinese majority (more than 50 percent but less than 60 percent of the composition) because Anwar would be immediately accused of trying to seek political legitimacy from Chinese rather than Malay voters.

(There's also a more practical reason - Anwar is more likely to win a greater percentage of Malay votes compared to Chinese votes so it is more politically advantageous to contest in a seat which has a Malay majority or plurality.)

Secondly, given the built in disadvantageous that competing in mixed seats entail, Anwar would not want to further disadvantage himself by going up against a cabinet minister, even one who has as many 'issues' which can be potentially exploited as Rafidah Aziz. It is likely that Lim Kit Siang choose to contest in Ipoh Timur in 2004 partly because the MCA incumbent there was not nationally prominent and did not have a position in the government.

Best options

So only Penang, Selangor and KL remain. The arguments for contesting in Penang are compelling ones. It is an ethnically mixed state and also happens to be Anwar's home town. Contesting there can be part of a larger strategy of co-operation with the DAP to try to make a serious dent in BN's power in the state.

BN's state leadership is expected to be in flux with the position of the next chief minister very much up for grabs within Gerakan and perhaps also within the BN, making this an issue which the opposition can possibly capitalise on. Finally, the issues of economic stagnation, price increases and corruption can possibly play well in this mostly urban state with an established history of voting for opposition voices to be heard in Parliament. Indeed, it is the only state which currently has at least one Malay, one Chinese and one Indian opposition representative in Parliament.

The only potential downside is that it is likely that Anwar will be 'forced' to contest in his seat of Permatang Pauh given the limited number of seats which are mixed and have a Malay majority/plurality. While standing there would strengthen PKR's chances of retaining this marginal seat, it does not create new electoral frontiers for the party in the way that, for example Kit Siang ( photo ) created by running in Ipoh Timor in 2004.

Nibong Tebal is certainly a possibility (44 percent Malay, 40 percent Chinese, 16 percent Indian) and was the only mixed seat in 1999 to vote in an opposition candidate (but since been changed by the 2003 delimitation exercise) as is Balik Pulau on Penang island (59 percent Malay, 36 percent Chinese, 5 percent Indian) but not Kepala Batas (Abdullah's seat) or Tasek Gelugor (77 percent Malay).

The risks for Anwar in Selangor or KL are greater but so are the potential rewards. He can break new ground for himself as well as for the party being the first Malay opposition representative at parliamentary level in either post-1969 Selangor or KL.

Voters in highly urbanised Selangor and KL have been the most affected by hikes in toll rates and petrol prices as well as the more general phenomenon of rapid urban inflation.

They are also more likely to respond positively to the messages of combating corruption (especially in the light of Abdullah's failure to carry out his election promises in this regard) and other issues which cross cut the ethnic divide, an area in which PKR can claim to have a comparative advantage.

The coat-tail effect of an Anwar candidacy in Selangor or KL should not be discounted especially if PKR, in co-operation with the other opposition parties, can field a strong slate of candidates in areas adjoining Anwar's seat. And there are more seats which Anwar could potentially contest.

A Gombak/Ampang/Pandan or a Kota Raja/Shah Alam/Kapar or a Subang/Kelana Jaya/Puchong strategy in Selangor is not out of the question. Neither is a Bandar Tun Razak/Wangsa Maju/Titiwangsa/Setiawangsa strategy in KL given the close proximity of these seats.

Evaluating a number of factors such as the strengths/weaknesses of incumbents, the potential to place strong opposition candidates in adjoining/surrounding seats, the presence of local issues which can be exploited, the organisational strength of PKR in these localities and so on, will be key in determining where Anwar can and should contest in Selangor or KL.

While contesting in Permatang Pauh seems the more obvious and safe option, Anwar can set a precedent for himself and his party in firmly establishing PKR's multi-ethnic credentials by winning a mixed parliamentary seat in either Selangor or KL.

Of course, all this is moot if Anwar is not allowed to contest at all. There's no substitute for actually having your name on the ballot. Even if he were allowed to campaign for other candidates or on behalf of his party, the effect would not be the same.

ONG KIAN MING is a PhD candidate in political science at Duke University.