Afr. i. polit. sci. (1997). Vol. 2 No. 2. 38-52 Voting Pattern and Electoral Alliances in Ghana's 1996 Elections Felix K. G.Anebo* Abstract In the 1996 presidential and parliamentary elections, Ghana's two major opposition political parties - the New Patriotic Party (NPP) and People's Convention Party (PCP) - which are traditionally very bitter opponents, formed an electoral alliance to defeat Rowlings and his National Democratic Congress. This paper analyses the factors that influenced the electoral alliance of the two traditional antagonists, and explains the reasons for their failure, in spite of their alliance, to win the elections. It argues, among other things, that changes in existing political alignments as well as voting patterns accounts for the electoral victory of Rowlings and his NDC in the elections. Introduction Common opposition to the Rawlings regime has made possible electoral cooperation between the country' s two major political antagonists, the New Patriotic Party (NPP) and People's Convention Party (PCP). In whatever way it is viewed, the electoral alliance between the NPP and PCP into what they refer to as "The Great Alliance" is politically very significant. The two parties come from two divergent political traditions and were historically bitter enemies. The PCP traces its ancestry to Kwame Nkrumah's Convention People's Party (CPP) which in its hey days was associated with radical nationalism, Pan-Africanism and socialism. The New Patriotic Party (NPP), on the other hand, derives its inspiration from Danquah and Busia's United Party, a strong advocate of liberal economic policies, protection of civil liberties and was generally pro-West. Up to now, some 1027-0353 © 1997 African Association of Political Science Voting Pattern and Electoral Alliances in Ghana's 1996 Elections 39 Nkrumahists blame Danquah/Busia followers for engineering the February 1996 coup that toppled the Government of President Nkrumah and with it, "Ghana's golden age". The banning of the CPP in 1969 and the subsequent disqualification of some of its leading members from engaging in party politics for ten years also bring bitter memories to many of Nkrumah's older followers. Danquah/Busia politicians in turn blame their colleagues in the Nkrumah camp for the excesses of Nkrumah's regime, in particular, the dreaded Preventive Detention Act (PDA) of 1958 which gave authority to Nkrumah's government to detain people without trial. Why then did the NPP and PCP decide to come together in an electoral alliance in 1996? In trying to answer this question I have structured my arguments into 3 parts. Part I is a brief examination of the traditional party cleavages that dominated party politics in Ghana before 31 December 1981. Part II examines the factors that precipitated the electoral alliance between the NPP and the PCP, and finally in Part HI the 1996 elections results and the emergent electoral cleavages are discussed. This paper is underpinned by an analysis of the complex linkages between party voting and issue voting at elections. It interrogates the thesis which attempts to understand recent electoral politics in Ghana from purely traditional party lines and contends that the Ghanaian society like all others is dynamic, and that the post independence electoral cleavages have changed significantly since 31 December 1982. Consequently, a clear understanding of electoral fortunes in the recent past must take cognizance of these changes. The twin concepts of party voting and issue voting developed by Nie, Verba and Petrocik (1989) provides the conceptual framework for this study. According to these authors, the main tenets of party voting are that a voter has a party identification. He or she must prefer one or the other of two parties. Such a preference must be a relatively long term one extending over several generations. In Ghana, party identification appears to be coterminous with voting intention; that is to say, if a person is going to vote for or has just voted for the candidate of a party he/she will almost always identify with that party. However, candidates must be from different parties, otherwise one cannot choose on the basis of party. The individual voter must use the party affiliation as the criterion for choice, though it is not always easy to tell whether this is the case. The notion of party voting, according to these writers, implies that a voter with no party identification cannot vote on party lines. Thus the growth in the number of independent candidates automatically reduces the number of voters who give a party their vote. The concept of issue voting is analogous to party voting, but somewhat more complicated. For instance, if there is an issue X and the two positions on that issue are Al and A2, the voter who prefers one of the two positions will vote for the candidate who holds that same position. Issue voting, therefore, according to Nie et al., requires: 40 F.K.G.Anebo i. That the individual must have a preference for either A1 or A2, (just as he ha to have a party identification for a party vote). Furthermore, just as part identification has to be more than a mere reflection of ho w one intends to vot or had just voted, so the issue preference of the individual must be more tha a mere reflection of his preference for the position of one candidate over th other; and ii. The candidates must offer a choice on the issues. One candidate must suppoi the other; or one must support one of the positions and the other(s) take n< position. Voters with issue positions can still vote on that basis: voters whi prefer Al will vote for candidate who prefers it, and those who do not prefe position A2 will vote for the neutral. But if both candidates support the sam position or are both neutral, issue voting is impossible [Nie, etal., 1989:71 Issue voting offers a number of complexities not found in party voting. Forom thing, issues do not always offer clear dichotomous choices of A1 versus A2 as i the choice between parties (Ibid: 250) There may be many possible positions 01 an issue. Sometimes the issue positions can be thought of as points along ; continuum. But if there were such a continuum, we could change the general ruh of issue voting to read: vote for the candidate whose position on the continuum i< closer to your own - implying that a voter has to be closer to one candidate thar another to cast an issue vote. The closer the two or competing candidates are to each; other, the more difficult it is it to cast such a vote and vice versa. A more complicated situation arises when there is more than one relevant issue Take for instance, a situation where there are two dominant issues in an election and each is equally important to voters. An issue voter would vote for the candidate who was closer to his position on the two dimensional plane defined by the two issues. The two issue case can be extended to a three or four issue case. For our present discussion what is important is the illustration this gives as to why issue consistency increases the potential for issue voting. Issue consistency converts the two dimensional space defined by the two issue continuum into one dimension Furthermore, issue consistency has a potentially important impact on the voting decision of the electorate. It not only simplifies the issues space for the voter, but it at the same time makes it easier for him to choose on the basis of issue position Thus, voters with a consistent issue position will be more sensitive to the issue positions of candidates. Traditional Electoral Cleavages in Ghana Ghana's traditional political division has been shaped, on one hand, by the body of ideas associated with Nkrumah - socialism, anti-imperialism, etc., and, on the other hand, by those of J. B. Danquah and K. A. Busia - political and economic Voting Pattern and Electoral Alliances in Ghana's 1996 Elections 41 liberalism, pro-West [Haynes, 1993: 451]. These two traditions produced corresponding patterns of electoral cleavages which to a large extent determined voting behaviour at elections. It is however true that between 1954 and 1956Nkrumah's CPP exploited divisions among their opponents and played skillfully on the alarm generated among southern Fante chiefdoms at the apparent resurgence of Ashanti nationalist ideology. According to Austin, this rather than ethnicity was the decisive factor in party affiliation and identification (Austin, 1964). But in general the nationalist leaders were able to divert inter-ethnic rivalries toward the colonial authority which was presented as the common enemy to all the tribes. Kwame Nkrumah and his associates were able to make the people believe that the colonialists had exploited them for far too long; and so there was the urgent need for all to come together to reclaim what rightfully belonged to them as citizens of (the then Gold Coast) Ghana. In post independence elections (and especially in the 1969 elections), the issues have been different. There is no longer an externl enemy. Instead, the old rivalries have emerged around personalities and the traditions they represented. Hence the 1969 election was, on the one hand, a personal struggle for the reins of government between Busia and Gbedemah and, on the other hand, a contest between the two political rivals (or traditions in Ghanaian politics) - the Busia/Danquah and Nkrumah traditions. It was believed that the Progress Party .which Busia led, was the revival of the former United Party - the party that had opposed the CPP government. Similarly, it was contended that the National Alliance of Liberals (NAL), led by K.A. Gbedemah, the man who is credited with creating the grassroots organization of the CPP, was the reincarnation of the disbanded CPP. Although new faces had appeared at the helm of the PP and NAL, the race for power was seen as a contest between the two political traditions. Beside this personality element, the PP directed its appeal to the regional groupings which the former opposition had relied on. It will be recalled that it has been in the Akan areas, and especially Ashanti that the opposition demand for federalism had crystallized against Nkrumah's unitary centralism during the 1956 election. In opposition to the CPP, Busia in 1954 had called for a revision of the constitution along federal lines, thus adding a national element to the agenda of the opposition and an intellectual gloss to the demand for federalism which the Ashanti cocoa farmers and the Northern Chiefs could hardly have devised by themselves (Austin, 1964: 29). In 1969, the alignment between more progressive Akans and the traditional leaders in the North formed the bases of the PP when we consider how the PP was able to win majority of seats in the Northern and Upper Regions. In contrast, the NAL's support was not exclusively regional. Although the Ewe support for Gbedemah increased as the elections drew nearer, his appeal was directed more towards the city voter. Analysis of the election results reveals that although NAL 42 F. K. G. Anebo parliamentary candidates won fifteen seats - Zeibila, Sandema, Bawku West, Ablekuma, Kpeshie, Dangbe, Shai, Ada, Yilo/Osudoku, Manya, Tamale, Gushiegu, Savelugu, Mion-Nanton, Gonja East outside the Ewe-dominated Volta Region, nineteen NAL parliamentary candidates forfeited their deposit in the Akan areas because they failed to secure the statutory minimum votes a candidate had to poll - one-eighth of the votes cast, to qualify for a refund of their deposit. Similarly, the Progress Party forfeited its deposit in six of the sixteen constituencies in the Volta Region. While the progress party won two seats - Nkwanta and the Krachi constituencies, in the Volta Region, the NAL did not win a single seat in the Ashanti, Brong Ahafo, Western and Central Regions, all Akan speaking areas. It is worth emphasising that even the two seats that the Progress Party won in the Volta Region were in the mainly Akan speaking enclaves (Nkwanta and Kete - Krachi) of the Volta Region. On the other hand, the only seats lost by the Progress Party in the Western Region were the two seats in Nzima, Nkrumah's own home district and Amanfi which were won by the Peoples Action Party led by Imoru Ayarna and P.K.K. Quaidoo of the All Peoples Republican Party respectively. In the Northern Region, the results were to a greater extent balanced as the voters divided their votes between the two major parties. Neither of the 2 dominant parties was identified with the North, even though the PP won a majority of seats. It was believed that local politics, rather than the Akan-Ewe ethnic cleavages, determined the results. Communal voting within lineage and chiefdoms (ethno-cultural factors) explain the electoral cleavages in the Northern regions. With no clear ethnic choice, nearly a a the seats were won with a minority or very close majority vote. For example the PP won 9 seats as against NAL's 5 seats. When the results of Tamale, Tolon, 'ewale, Nanumba, Savelugu, Gusheigu and Yendi are compared this comes out quite clearly. n Greater Accra, the varied socio-economic and political backgrounds of voters produced a divided vote in the 9 constituencies, so that the PP and NAL both captured 3 seats each, the United Nationalist Party won 2 seats and an independent candidate, Harry Sawyer, 1 seat. But even here the election results portrayed the ( political cleavages as well as a voting pattern which was along ethnic lines. th ,^Otller d°minant pattern worth noting is the rural-urban divide. For instance AL appeared to have attracted voters between the traditional-modern ex- e m^s " mamly Ghanaians who appeared to be more urbanized or at least subject c° Uf an '"fluences because of their social environment. As one moved across the o f in \ 0 ln vote1^ f r ° m thC N ° r t h WeSt t o w a r d s BolSatanga and Tamale in the North, the PP CS tended to drop while the NAL picked up more votes, indicating the influence AhSCfClal a n d s p a t i a l m o b i l i ty uP°n voter preferences. Even in Ashanti and Brong t ° r e 8 l o n s ' t h e v o t e s for PP dropped in the capitals. Although some of the shifts g pattern can be explained by the presence of enclaves of Non-Akans, Ewes Voting Pattern and Electoral Alliances in Ghana's 1996 Elections 43 and people from the North in these capitals, ethnic concentrations are not enough to account for the voting pattern. A third, and equally important voting pattern was that of party voting based on the two dominant party tradition - the Danquah /Busia and the Nkrumahist traditions. Emerging Cleavages in the 1979 Elections A rigorous analysis of the 1979 elections reveals 4 broad patterns of electoral cleavages and voting behaviour. The first conspicuous pattern is that unlike the 1969 elections, there was no real difference between urban and rural voting. The urban voting pattern followed the same trend as those of the rural areas. The PNP enjoyed popularity all over the country compared to the other political parties, and for that matter independent candidates. Dr Limannn and his PNP won 73 of the 140 constituencies in the first ballot; and in the second round 105 as against 35 by Mr Victor Owusu of the PFP. Furthermore, whereas the PNP was able to win at least a constituency from each of the nine regions of the country, the other parties including the independent candidates did not achieve such an enviable feat. An important voting pattern which is similar to what prevailed in the 1969 elections, is party voting: the electorate voted along the lines of the Danquah/Busia and Nkrumah traditions. Although Dr. Limann was able to win majority votes in the Akan and the non-Akan areas, block voting in areas like Nzema East, Cape Coast, Walewale and the Akan constituencies in the Volta Region depicts voting along traditional party lines. Historically, all these areas were a stronghold of the CPP, and Dr. Limann's PNP benefited from the popularity of the CPP tradition in these areas where Kwame Nkrumah's CPP had won all elections in this country except the elections of the 1969 when Nkrumahists were banned by the ruling National Liberation Council (NLC). 1996 Elections and the Great Alliance (GA) Several reasons could be adduced to explain why today's followers of Danquah Busia and Nkrumah buried their past and present differences, and formed a united electoral alliance to challenge Fit. Lt. Rawlings on 7 December 1996. The first reason is the enormous political popularity of Fit. Lt. Rawlings especially in the rural areas of Ghana. Any objective analysis of contemporary Ghanaian politics would conclude that Fit. Lt. Rawlings remains the most popular politician in the country today. Having ruled Ghana for the past fifteen years, he enjoys all the political benefits of office holders in Africa, including access to vast state resources for political campaigning, patronage, and access to influential local notables nation-wide. Consequently, the opposition parties, especially the NPP and PCP reasoned that without a united front, Rawlings and his party, the NDC would easily win the December 1996 elections. The only way to prevent that from happening was therefore to unite their supporters and bring the bulk of the electorate behind 44 F. K. G. Aneho them instead of diffusing their potential and actual electoral support, and risk loosing the elections as happened to them in 1992. Apart from Fit. Lt. Rawlings' popularity, the large floating voters among the Ghanaian electorate also emphasised the need for an opposition electoral alliance. Opinion polls and surveys during the electioneering campaign placed Ghana's floating voters at approximately 45 per cent of registered voters. Unattached to any political party in particular, floaters may find a disunited and fragmented opposition a big disincentive to vote if they strongly wanted a change of government. Some floating voters may even decide to throw their weight behind the incumbent NDC government led by Fit. Lt. Rawlings as a solution to the state of confusion and despair in which they may find themselves. The Kumi- Preko demonstrations organized by the Alliance for Change (AFC) in Accra, Kumasi, Koforidua, Tamale and other regional capitals against the introduction of the Value Added Tax (VAT) and the extremely high level of popular participation in them also acted as a catalyst for an opposition electoral alliance. What is more, the AFC, which gained its popularity from the Kumi- Preko demonstrations drew its most prominent leaders from both the NPP and the Nkrumah political traditions. The success of those mass protests and the subsequent withdrawal of the VAT did not only show the size of Ghana's floating voters and the possibility for a credible opposition party to pool them along, but also underscored the need for unity among the opposition forces if they hoped to succeed in the electoral contest with Rawlings and his NDC. It lent credence to the saying that "in unity lies strength". Indeed, the possibilities for a successful united front by the opposition were so powerful: if Kwesi Pratt, Charles Wereko-Brobby, Akuffo-Addo and others from the CPP and the UP families could agree to organize such historic demonstrations, nothing should prevent the two traditional political rivals from coming together. Indeed, the opposition parties had no reason to allow past political animosities to stand in their way to harvesting the potential electoral success that the political situation promised. In the circumstances, an electoral alliance of the main opposition parties became the greatest imperative of the moment. The force of example from other African countries, especially Zambia and Benin, where a united front of opposition parties had facilitated the defeat of the incumbent party made the need for unity equally urgent. The drive for unity was further strengthened by the argument that the ruling NDC itself derives its strength from political alliances - in 1992, the Progressive Alliance of three parties comprising the NDC, NCP and Egle; and in 1996 comprising the NDC, Egle and the Democratic Peoples Party [DPP]. Not only has the NDC carefully constructed alliances with new political parties; some of its leading members such as Mumuni Bawumia, Kojo Tsikata, Obed Asamoah, A. A. Munifie, Fustina Nelson, just to mention a few, were drawn largely from the Kwame Nkrumah and Danquah/Busia political traditions. Voting Pattern and Electoral Alliances in Ghana's 1996 Elections 45 Finally, the need for an opposition electoral alliance between the Danquah/ Busia and Nkrumah families was rationalised on historical grounds. Historically, it was argued, the CPP came from the same nationalist stock as the UGCC (the Danquah/Busia family). As an offshoot, it took away its name, platform and many others, including the use of the name - Ghana for independent Gold Coast. It was Dr. Danquah who drafted the long telegram to the Secretary of State for the Colonies on 29 February 1949 in which the Convention called for the appointment of a Special Commissioner to assume the powers of an interim government pending the convening of a Constituent Assembly. He backed it up with the article "Nananom, the Hour of Liberation has Struck." published on 23 March 1948. It would be strange if he and all the Convention leaders, after all this, did not want immediate self-government as some would argue. It was true that in the end UGCC leaders contented themselves with the Coussey Committee which produced the 1950 Constitution, dubbed as "bogus and fraudulent" by Nkrumah and his CPP. It could be seen, on the other hand, that when in protest, Nkrumah organised the Positive Action campaign against the British colonial administration, he came back to work under the governor who had jailed him, using as a basis that "bogus" Constitution, until he produced an improved one in 1954. And in the 1951 elections, when Nkrumah won 38 seats in the Legislative Assembly and Danquah and his allies won barely 3 seats, Danquah wrote in a letter dated 13 February 1951 to Nkrumah, a day after his release from prison: My Dear Kwame, This is a glad occasion. You have fought the good fight and triumphed for the justice of our cause. Your imprisonment and your release are symbolic of the conquest over imperialism. You may have made mistakes, as even the greatest do, but you have passed through a baptism of fire, a spiritual fire and you have suffered bodily in the cause of our Motherland ... May the Gods of Ghana help and guide you and may the baseless misrepresentations and misunderstandings of the past that engendered disunity in our struggle be buried with their own past. We started with a United Gold Coast. Let us complete the work for united motherland ... May God bless you." These facts, the argument goes, affirms the view that the two nationalist parties wanted immediate self-government; but they were also realists and knew when to change gear. Thus, adherents to opposition electoral alliance argue that these two party traditions in some respects can best be represented by the Ghanaian version of Tweedledum and Tweedledee. In the "Adinkra" symbolism the two party traditions are "Funtumrefu-Denkyemrefu" or twin crocodiles with a common stomach who nevertheless quarrel over food and yet have much in common in terms of political demands and aspirations. The emphasis was on commonalties which derive, at least, in part from a common historical experience. 46 F. K. G. Anebo The 1996 Elections and Emerging Political Cleavages Table 1: Presidential Elections Results PA) 7,225,161 4,099,760 100.0% 57.4% 39.6% 3.0% 2,825,715 2,10,980 Total Votes cast J. J. Fit. Lt. Rawlings (Progressive Alliance — J. A. Kuffuor (Great Alliance — GA) E. N. Mahama Table 2: Parliamentary Elections Results Number of seats won 133 61 5 1 Party NDC (Led by Rawlings — PA) NPP (Led by Kuffour — GA) PCP (Supported Kuffour — GA) PNC (Led by Mahama) TOTAL 200 As shown in Tables 1 and 2, President Fit. Lt. Rawlings and his NDC won both the presidential and parliamentary elections overwhelmingly. The results of the presidential election gave President Rawlings a clear victory. He won 4,099,760 votes out of a total of 7,225,161 valid votes cast, (or 57.4 per cent) compared to 58.3 per cent votes obtained in the 1992 elections. J. A. Kuffuor of the G. A. won 2,825,715 (39.6 per cent) of valid votes cast. This was a better performance in comparison with Adu Boahen's in 1992 when the latter obtained 30.4 per cent of total valid votes cast. Two factors may explain this improvement in Kuffour's performance: (i) the formation of the opposition alliance (the G. A.), and (ii) the personal qualities of J. A. Kuffuor himself. A cursory look at both presidential results shows that the PNC performed woefully. E. N. Mahama won just 210,980 or 3 per cent of total valid votes cast, a drastic drop, when compared with about 7 per cent obtained by his predecessor, Dr Hilla Limann in 1992. The implication is that Limann is more popular than E. N. Mahama. In the parliamentary election, the NDC won 17 of the 21 seats in Brong -Ahafo, 15 of the 26 seats in Eastern region, 18 of the 23 seats in the Northern region, and 12ofthe 19 seats in the Western Region. The NDC won all the 19, 12and8seats in the Volta, Upper East and Upper West respectively. It was only in the Ashanti region that the NDC performed poorly: it won only five of the 33 seats, whilst its closest rival, the NPP/GA won the majority of seats. It is remarkable that unlike the opposition political parties, the NDC won constituencies in each of the 10 regions making it the only party with the broadest national support. Voting Pattern and Electoral Alliances in Ghana's 1996 Elections 47 Implications for Subsequent Elections The 1996 general elections can be said to have consolidated the transformation of the political topography of Ghana, probably for all time. As I have argued elsewhere in this paper, for most of this country's history, political life has been dominated by the two political traditions - the Nkrumah CPP tradition and the Danquah/Busia tradition which divided the electorate into 2 broad voter camps. Before 1992, the 3 elections which were held to usher the country into constitutional rule had been won by parties from either of the two traditions. Parties of the Nkrumah tradition had won twice, in 1957 and in 1979, and those of the Danquah- Busia tradition once - in 1969. From 1981, when. Fit. Lt. Rawlings launched his 31 st December Revolution, to the 1996 elections Fit. Lt. Rawlings has succeeded in establishing acredible "Third Force" in Ghanaian politics. The presidential and parliamentary elections of 7 December 1996 marked the second time in two successive elections since 1992 that parties which claim their heritage from Nkrumah and Danquah/Busia (Ghana's nationalist leaders who ruled immediately after independence), have lost power to the National Democratic Congress (NDC). In the words of Hudu Yahaya, General Secretary of the NDC: The post-independence part of our history ended in 1970. The arrival in 1992 of the NDC introduced a new epoch - a new thinking - very different from what prevailed after independence through the early 70s. {Ghanaian Times, 19 December 1996). Charles Wereko-Brobby, who resigned from the NPP to found the United Ghana Movement (UGM), seems to agree with Hudu Yahaya when he states: The opposition underestimated the power of Fit. Lt. Rawlings. It is now the leading political force and it is only when they (the opposition) realise this that they can begin to appreciate the extent of the problem they face (Ghanaian Times, 19 December 1996). While the issue of a "Third Force" in Ghanaian politics is indisputable, some analysts point to the ethnic factor in both the 1992 and 1996 elections in explaining the electoral fortunes of the NDC. Undoubtedly, the results of the presidential and parliamentary elections of 1996 portrayed an entrenched pattern of voting in two of the country's ten regions. This pattern most closely reflected that of the 1992 elections. For example, the highest vote for any single constituency in the presidential elections came from Bantamain the Kumasi metropolis of Ashanti, where Kuffuor got 81,128 votes as against 18,487 for Rawlings and 1,076 for Mahama. In the 1992 Presidential elections, Rawlings - the NDC presidential candidate won 93.2 per cent of the votes in the Volta Region, and all the seats in 48 F. K. G. Anebo that region except Nkwanta constituency. It swept all the seats in the three Northern Regions, whilst all the constituencies in Ashanti and the main municipalities and metropolis voted for the NPP. Similarly, in the 1996 elections the NDC presidential candidate, Fit. Lt. Rawlings won 94.5 percent of the votes cast in the Volta region and swept all the 19 seats in that region, 18 of the 23 seats in the Northern Region, and all the seats in the Upper East and Upper West Regions. As in 1992, the NPP presidential candidate, J. A. Kuffuor also won 65.8 per cent of the votes cast in Ashanti and 27 of the 32 seats in that region. Most of the votes from urban areas such as Sekondi, Cape Coast, Tamale and Koforidua went to the NPP. An interesting development however is that while Fit. Lt. Rawlings won in the presidential elections in constituencies, such as Ablekuma south in Greater Accra; Offinso North and Asokwa East in Ashanti; Wenchi East in BrongAhafo; Bimbila and Wulensi in the Northern region, and Ellembelle in Western Region, his party's parliamentary candidates were defeated in these same constituencies due most probably to local political and individual personality factors (Ayee, 1997: 18). This pattern of voting has been explained by the ethnic factor. While the ethnic factor in the polls cannot be totally dismissed, particularly with reference to the Ashanti and Volta regions, it must however, be pointed out that the voting behaviour in the two regions is not representative of voting behaviour in the rest of the country. We must therefore look elsewhere for an explanation and understanding of the electoral politics in the 1996 elections. As stated earlier, to understand the electoral politics in the 1996 elections, one must take cognizance of the re-alignment of political forces in the country as well as the movement away from mainly "party voting" to a mixture of "party voting" and "issue voting" which occurred among a large segment of the electorate. Simply put, new political cleavages have emerged or are in gestation among the Ghanaian electorate. Evidence across the country from the election results and subsequent political developments after the elections point to the fact that the bulk of what has been the traditional CPP supporters and sympathisers had joined forces with the NDC in a new centre left coalition which is the Progressive Alliance. This was evidenced by the fact that the NDC captured the Eastern and Brong-Ahafo Regions, previously thought to be strongholds of the NPP (Danquah/Busia tradition) and Western and Central Regions which were seen as the preserve of the CPP tradition. An apt description of the political cleavages in Ghana today may be stated as follows: (i) the centre-left occupied by the NDC and the Progressive Alliance; (ii) the centre-right represented by the PCP and other parties of the Nkrumah political tradition; and (iii) the right occupied by the NPP of the Danquah/Busia political tradition. Voting Pattern and Electoral Alliances in Ghana's 1996 Elections 49 Indeed, the NDC has virtually appropriated the political position of the CPP tradition while the latter, represented by the PCP and others, have moved to the centre-right. Several reasons could be adduced for this restructuring of political alignment in Ghana. The first plausible explanation for such re-alignment is the close affinity between the CPP's radical and revolutionary ideas and that of the NDC's populism. Furthermore, to all intents and purposes, the social democratic aims and objectives of Rawlings and his NDC are in tandem with those originally epitomised by the Nkrumah tradition. It would be recalled that by the close of the 1970s the Ghanaian state faced a series of crises. These crises led to widespread political agitation by a large segment of the populace. By exploiting such crises to capture political power, the Provisional National Defence Council (PNDC) inserted itself in that rather volatile situation, and so it had no choice but to deal with them as a condition for its own survival and legitimacy. As a first step, it initiated massive mobilisation of the discontented strata of the Ghanaian society - the lower middle class, working classes as well as the unemployed, semi-employed and informal sector operators in a political struggle against the Ghanaian establishment which comprised the upper and middle classes in both the private and public sector [Ninsin, 1996: 27]. This was the genesis of the re-alignment of political cleavages in current Ghanaian politics. Describing these emergent political cleavages Mike Oquaye writes: The revolution initially divided the nation into "people" and "citizens" Those who had a role to play in state affairs and those who did not. Citizens included the professionals, managers, and the business community, landlords; prosperous farmers and bourgeois intellectuals; while "people" included the underprivileged, downtrodden, workers rank and file; police men and soldiers, small time fanners, petty traders and progressive intellectuals/ students [Oquaye, 1995: 263]. The people's cadres were from varied social origins and traditions and successfully mobilised political support for the NDC. They were concentrated in organisations such as the Progressive Voluntary Organizations (PVOs), Mobisquads, ACDRs, Liberated Nkrumahists Brigade (LNB) of the NDC and the 31 December Women Movement (DWM). Common to these groups is their revolutionary zeal. Most of them attained political maturity under the reign of Fit. Lt. Rawlings and therefore know no other politicians, style of politics, strategies and achievements except that of Fit. Lt. Rawlings. Nkrumah, Danquah and Busia are not names that strike a familiar political cord in them. A second re-alignment is the increased relationship between policy issues and voting choice and a decreased relationship between traditional party lines and voting choice. A post- election survey conducted by the Department of Political 50 F. K. G. Anebo Science of the University of Ghana supports this view. For instance, out of a sample survey of 2,700 voters who were asked as to whether they identified themselves with any political party, 80percent indicated that they are affiliated with a political party of a sort. When respondents were subsequently asked why they voted for a particular party, 65 per cent said they voted for a particular political party because the party had good policies that would ensure the development of the country. When this is contrasted with the alternative answer as to whether they voted on the bases of the traditional party affiliations, only 2.4 per cent of the respondents said they voted according to traditional party considerations. Thus, the contrast between the attitude/vote correlation and party/vote correlation was dramatic in the 1996 elections. When respondents were asked to rank issues they considered most important which politicians should address, the responses were as follows: out of the 2,700 respondents, 1,209 (44.8%) ranked unemployment; 1048 (38.8%) ranked inflation as an important national issue while 127 (4.7%) ranked corruption . On the economy in general, most respondents were quite pessimistic. When respondents were asked about the state of the Ghanaian economy on the eve of the elections, 19.5 percent said the economy was in a very bad state; 16.7 per cent said that the economy had become worse since 1992 while 13 per cent of them said the economy had become better. Respondents were further asked whether the quality of their lives had improved since the 1992 elections. A striking 60.3 per cent said "No" while 32.2 per cent answered "Yes". When this is contrasted with the G A campaign slogan of "Hwe wo asetena mu na to aba no" (translated as "You should reflect on your economic situation before you cast your vote") it becomes difficult to understand why the electorate voted for the NDC (and the Progressive Alliance) rather than the NPP and the GA. The explanation however lies in the high correlation between consistent issue position of the electorate and that of the NDC. It would be recalled that the NDC had built its campaign around the issue of "Continuity, Stability and Development". It would seem that most eligible Ghanaian voters had adopted the philosophical position that "The devil you know is better than the angel you do not know". Accordingly when respondents were asked to indicate what issues influenced their vote, 41.9 per cent said it had been influenced by good policies of the NDC; 15.7 percent said it had been influenced by the good and honest leadership of Rawlings. The implication then is that Rawlings and his NDC won the elections on the strength of the good policies the NDC government under his leadership had pursued in the past. The bottom line is now clear. The NDC has come to stay in Ghanaian politics. By building a hierarchical organic party structure from the national to the constituency level, it succeeded in wining constituencies in each of the ten regions of the country. This underscores the fact that the NDC is better organised than the opposition political parties. The construction of a corps of cadres comprising the Voting Pattern and Electoral Alliances in Ghana's 1996 Elections 51 Committee for the Defence of the Revolution (CDRs), the Patriot Club, DWM, PVOs and others greatly facilitated the mobilisation and dissemination of the party manifesto in the rural areas. It is not surprising therefore that there was a high rural turnout for the party in both the 1992 and 1996 elections. The failure of the GA to wrestle power from Fit. Lt. Rawlings and his NDC clearly shows that there is no hope for an effective opposition electoral alliance. The feet-dragging that characterised the signing of the GA pact coupled with the post-election differences between the PCP and NPP parliamentarians on the issue of vetting of cabinet ministers from Rawlings' previous government leaves much to be desired. Alliances are good only if political forces who broker them share the same ideological persuasion [Ayee, 1996]. This development has significant implications for multiparty democracy in Ghana. There is no need belabouring the point that to win elections in any African country today requires enormous financial and other resources. Considering the weak financial position of the opposition political parties compared to the NDC, it is doubtful whether any of the opposition parties could compete favorably with the NDC without some form of state funding. If the issue of funding and the need for a level playing field for all registered political parties are not seriously addressed, Ghana would remain a de facto one party state for some time to come. This will naturally deprive the country of the advantages that accrue from a multiparty democracy. In addition, the danger of returning the country to the authoritarian regime of the recent past cannot be ignored. As observed by Luckham, democracies will remain at risk so long as the manifold legacies of authoritarian rule are not confronted, including privileged, non-accountable military and security bureaucracies [Luckham 1996: 1]. Note * Lecturer, Department of Political Science, University of Ghana, Legon. References Ayee, J. R. A., (1997) "Ghana's 1996 general Elections: A post-mortem" AAPS Occasional Paper Series, Vol. l,No. 1. Dennis, A., (1964) Politics in Ghana 1946-60 Oxford University. Giovani, S. (1970) "Concept Misinformation in Comparative politics" American Political Science Review, Vol. LXIV, No. 4. Haynes, J (1993) "Sustainable democracy in Ghana? Problems and prospects" Third World Quarterly, Vol. 14, No. 3, 1993. Luckham R., (1996) "Democracy and the Military: An Epitaph for Frankenstein's Monster?" Democratization, Vol. 3, No. 2. Nie et. ai, (1989) "Parties and Elections" in Randall B. Ripley and Elliot E. 52 F. K. G. Anebo Slotnick (eds.): Readings in African Government and Politics, Donnelley and Sons. Ninsin, K. A. (1996), "Ghana: Beyond Crisis and Adjustment" Africa Development Vol. XXI, Nos. 2 & 3. Oquaye, M, (1995) "The Ghanaian Elections of 1992 - A Dissenting Vie w" African Affairs 94. Newspapers/Documents The Government Gazette Election Results. The New Africa June 1979. The Legon Observer, Vol. XI, No. 10, 13-26 July 1979. The Legon Observer, Vol. XI, No. 10, 13-26 July 1979. The Legon Observer, Vol. XI, No. 11, 27-10 August 1979. West Africa, June 1979.