GHANA ELECTIONS: FINAL OPINION POLL PREDICTS A

C O N F I D E N T I A L SECTION 01 OF 02 ACCRA 001538 S U S P E C T E D D U P L I C A T E SIPDIS DEPT FOR AF/W E.O. 12958: DECL: 12/04/2028 TAGS: PGOV PHUM PINS PREL KDEM GH SUBJECT: GHANA ELECTIONS: FINAL OPINION POLL PREDICTS A RUN-OFF REF: ACCRA 0473 ACCRA 00001538 001.2 OF 002 Classified By: POLCHIEF Gary Pergl for reasons 1.4 B and D 1.(C) SUMMARY. Ben Ephson, independent pollster and newspaper editor, predicts a presidential runoff following the December 7 election. Ephson's "The Daily Dispatch" predicts the ruling New Patriotic Party will get between 48.2 and 50.2% of the vote, with the opposition National Democratic Congress (NDC) polling between 44.7 and 46.7%. His surveys show the parliamentary race tightening, with significant losses for the NPP and some gains for the NDC. He projects NPP with 106 definite seats (they currently hold 128 seats) and NDC with 102 seats, up 8 from their current ¶94. But with 21 races too close to call, either party could win a majority in Parliament. He predicts that Ghana's other opposition parties, the Convention Peoples Party and the Peoples National Convention, will lose ground, with the CPP facing the possibility of no representation in the next Parliament. Ephson lamented the amount of money flowing into the campaign, particularly by the NPP, and again told EMBOFFs that he had been offered a large bribe by an NPP supporter to alter his polling data to show the party winning in the first round. END SUMMARY. ¶2. (U) In a final pre-election conversation, Ben Ephson met with the political section on December 3. In a previous meeting on November 14 (reftel),Ephson predicted a run-off election for President, a position he continues to support. Ephson's polling methods involve "person in the market" interviews in each of the country's 230 Parliamentary constituencies, and are probably as statistically robust as other methods used. With a purported margin of error of 2 percent, Ephson's poll, completed in late November, shows the NPP running at between 48.2 and 50.2%, the latter just sufficient for a first round victory. The poll shows the NDC running at between 44.7 and 46.7%, in the range of its support in the past two general elections. The third place CPP runs in the range of 5.3 and 7.3%, with the other parties polling, combined, at less than 2%. Ephson conducted three surveys during the election season, each polling a total of 4060 registered voters from each of Ghana's 230 parliamentary constituencies. The data was weighted by the number of registered voters in each constituency. 3.(U) The November polling shows expected patterns of regional voting. The NPP polls at 79% in its heartland, Ashanti, while the NDC runs a strong 70% in its stronghold of Volta. The poll shows the parties very close in Accra, at 42% for the NDC and 45% for the NPP, a change from 2004 when the NPP won the region by 52-46%. The NDC continues to lead in the three northern regions of Upper East, Upper West and Northern, although these areas have low overall vote counts, compared to the southern regions. The poll shows the NPP holding a strong lead in Brong-Ahafo (63% to 29% for the NDC) and in Eastern (63% to 33%). Central and Western Regions show leads for the NPP over the NDC that are similar to Greater Accra (46 to 42%). 4 (C) Ephson lamented the role of money in the current campaign, saying that he "had never seen so much cash before." Again, he said quite openly that the NPP had tried to bribe him with at least $20,000 to produce a poll favorable to them-- he named the party official offering the money as Gabby Ochere-Darko, a cousin of the party's candidate, Akuffo- Addo. He also said, although without presenting the evidence he claimed to have, that the NPP was trying to get the names of polling agents for other parties. The NPP would use this information to bribe agents, and was willing to pay $1,000 for polling agents to collaborate with vote rigging. Ephson said that in the past six weeks, in ten swing constituencies, the NPP had spent large sums of money trying to secure the parliamentary seats. Asked by POLOFF where the money was coming from, Ephson said it was coming out of the government's coffers, primarily from kickbacks on government-awarded contracts. ¶5. (U) Ephson gave POLOFFS detailed region by region rundowns of specific parliamentary races. In Greater Accra's 27 constituencies, he predicts the NDC will gain 4 seats at the NPP's expense, with two races too close to call. Of Ashanti's 39 seats, the NPP will fall from 36 seats to 35, with two races too close to call. In Central Region, with a total of 19 seats, NPP will lose 3 races to fall from 16 to 13 seats, and NDC will gain one. CPP will lose its only seat there, and 3 seats are too close. Of the Northern Region's 26 seats, NDC will gain 5 for a total of 22, NPP will lose 6 to fall from 8 to 2, and 2 races are too close to call. Brong Ahafo Region will basically remain the same, with 14 NPP seats and 10 for NDC, but 2 races are too close. NPP now ACCRA 00001538 002 OF 002 leads NDC 22 to 6 seats in Eastern Region, but 3 NPP seats and one NDC seat are too close, giving NDC an edge to pick up a few seats there, In Western Region, current distribution is NPP 13, NDC 7, and CPP 2. NPP should hold firm there, but both CPP seats and one NDC seat are too close. In the NDC's stronghold Volta region, the only NPP seat will go to the NDC this time, giving the NDC a sweep of all 22 seats, with the possible exception of an NDC member running as an independent. For the 13 Upper East seats, NDC now holds 9, NPP 2, and the PNC 2. Ephson projects the NDC will pick up at least one seat from the NPP, and the remaining NPP seat and both PNC seats are too close to call. Finally, in Upper West, NDC gains one seat at NPP's expense, and PNC retains its 2 seats. ¶6. (U) When you add up the numbers, the NDC, according to Ephson, has a net gain of 8 seats, from 94 to 102. The NPP loses 22 seats, going from 128 to 106. He says that 21 constituency races are so tight that he can not make a prediction. Depending on how those contests break, either party has a shot at a majority in Parliament. Ephson sees both minor parties losing ground in Parliament, with the CPP facing the prospect of legislative oblivion. ¶6. (C) COMMENT: Although Ephson's methodology does not, in our minds, lend itself to his claim of a 2 percent margin of error, his statistical sampling is large enough to give his polls some credence. More importantly, Ephson is not in anyone's pocket, which is not something that can be said about every other opinion pollster in the country. By default, he is the best game in town. Although he said that he hopes the NDC wins, and he fears the increasing power of the NPP, he admitted that he voted CPP in the limited early balloting that took place December 2. We believe him when he says that his ultimate meal ticket is his long-term credibility, and that despite numerous offers, he has never taken money from any political party. (He did, however, accept $40,000 in the form of a Democracy and Human Rights grant from us, and we suspect he gets support from other donor partners and NGOs.) In this final week before elections, Ephson has become something of a media darling, in near constant demand on radio and television, because he is probably the most trusted election forecaster in Ghana. Four years ago, Ephson's predictions came closer than any other poll's to the final tally, but he still underestimated NDC's vote by 6-8 percent. He thinks he will be closer this year, and he tried to extract the promise of another grant if his predictions fall within his margin of error. TEITELBAUM