Kenya: Elections still too close to call (and implications)

Kenyans headed to the polls on Tuesday to elect a new president, parliament and county governors (totalling 1,882 posts in all). At the time of writing the race is still too close to call, with local media houses The Standard showing a 49.78% to 49.55% margin in favour of William Ruto with just over 30% of the votes left to tally and the Daily Nation and Citizen respectively showing a 50.41% to 48.91% split and 49.48% to 49.12% split in favour of Raila Odinga with just under 20% of the votes left to count.

These preliminary results were tabulated based on the official “Form 34A” submitted for each constituency and posted on the website of the Independent Electoral and Boundaries Commission (IEBC). The results differ not only from the pace at which each organisation is tallying the results, but also due to a degree of randomness in the sequence of constituencies that they tally, although with The Nation and Citizen having counted more votes it appears Odinga may have a slight edge at the moment.

The IEBC has until 16 August to declare a winner, but hopes to make an official announcement earlier than that. While it has already received 99.9% of the “Form 34A” sheets in digital format, it is still waiting on the physical delivery of many of the forms, which is necessary to officially verify the results. As such, the winner is unlikely to be announced today and counting is likely to stretch into tomorrow and possibly into the weekend.

Under Kenya’s electoral law, a presidential candidate needs to secure more than 50% of the valid votes cast nationally and at least 25% of the ballots cast in half of Kenya’s 47 counties to be declared the winner. If no candidate meets that threshold, a run-off election between the top two candidates must be held within 30 days. Based on how close the race is between the top two candidates the chance of a run-off remains high, despite the two other candidates receiving less than 0.7% of the vote.

If neither candidate is able to surpass the thresholds to avoid a run-off, a fresh election between the top two candidates will be held within 30 days. If a candidate wins in the first round, they will be sworn in within 14 days of the announcement. However, any citizen or group can submit a challenge to the supreme court within 7 days of the official announcement of results, which will then have 14 days to issue a final (unappealable) judgement. If the results are annulled, a fresh election must be held within 60 days of the ruling.

This is exactly what happened in 2017, when Odinga challenged his loss and the Supreme Court annulled the results, triggering a fresh election. With a razor-thin margin compared to 2017 (when Kenyatta won by 9.3%), whoever loses is likely to challenge the results. However, the IEBC has learned from its 2017 mistakes (missing 34A forms, which the IEBC has transparently posted to its website this time around, were a key motive for the annulment), and the brutal murder of the IEBC’s ICT head before the 2017 elections lent further creditability to the tampering allegations, so the results are less likely to be overturned over irregularities this time around.

Risk of violence

A key concern leading into elections was the risk of violence, with post-election violence in 2007-08 leading to an estimated 1,200 deaths and displacement of up to 600,000 Kenyans from their homes. The 2013 and 2017 elections were notably less violent (with a still notable but much smaller 50 deaths estimated in 2017), and the devolved political system ushered in under the 2010 constitution makes for a less winner-takes-all environment, reducing the risk of a repeat of the horrible violence of 2007-08.

While both candidates have promised to accept the results, Odinga’s remarks have been more qualified than those of Ruto, saying in an interview that "If I lose fairly, I will be the first to concede defeat. But if it is not fair, then I will follow the normal channels to address the issues." Further, with this year’s election marking Odinga’s fifth attempt at the presidency, he is unlikely to go down without a fight if he loses again.

This sentiment is bolstered by Odinga’s high degree of confidence in his victory, with Odinga leading by a 46.7% to 44.4% margin in the final pre-election poll by Tifa Research with around 5% of voters still undecided and another 2% declining to share their preference. Indeed, Odinga said in a recent interview that “I’m certain that if it’s done fairly, there’s no way that I can lose this election,” and has rejected the results of his last three attempted presidential runs in 2007, 2013 and 2017.

Regardless, either candidate is likely to challenge the results in court. But with limited reports of irregularities and a commendably transparent electoral process run by the IEBC, the Supreme Court is likely to uphold the results. Further, the Supreme Court has bolstered its credibility in recent years via its 2017 election annulment and more recent decision to throw out outgoing President Kenyatta’s initiative to amend the constitution, making it more difficult to refute its rulings.

Overall, we think Ruto is more likely to accept the outcome if he loses, limiting the risk of violence. Conversely, if Odinga loses he will likely feel cheated and there is a risk that he makes a final push for the presidency by stirring up the anger of his large and enthusiastic support base. As such, while the process has so far been peaceful and widespread violence is not our base case regardless of the outcome, the process still needs to play itself out before the nation can breathe a sigh of relief.

Policy implications

While we think Ruto is the marginally more market-friendly candidate, there didn’t appear to be much of a policy difference between the two candidates on the campaign trail. Ruto framed the election as “Hustlers versus Dynasty”, contrasting his humble beginnings selling chickens with the dynastic political lineage of Kenyatta and Odinga (both of whose fathers were key figures in Kenya’s drive towards independence and early post-independence politics), but ultimately both are establishment candidates.

Ruto’s campaign centred on scaling back mega projects to refocus support to small businesses and farms (including investing KES500bn in agriculture and SMEs over 5 years, subsidising farm inputs, and launching a “Hustler Fund” to provide KES50bn of credit a year to small-scale traders). He also promised to limit borrowing outside priority areas like agriculture with a high potential for job creation, digitise tax mobilisation to double VAT collection and widen the tax base to increase revenue collection by 70% without hiking tax rates, provide interest-free student loans and double funding for lending and research, and reduce government corruption.

Odinga’s platform focused on many of the same policies, also promising to increase subsidies on agricultural inputs and boost support for small businesses (albeit with tax relief rather than new spending/investment) and emphasising the need to back spending on “white elephant” projects like the Standard Gauge Railway (which has faced widespread criticism of late both domestically and abroad – an issue we flagged two years ago – see here and here). However, Odinga plans to redirect those savings towards spending on new (and presumably smaller) road, rail and sanitation infrastructure and a more aggressive expansion of the social safety net compared to Ruto.

Odinga’s social agenda centres on the provision of universal healthcare, free education through the tertiary level, a reduction of the fuel excise, and new cash transfers of KES6,000/month to poor households. Odinga had also planned to more than double budgetary allocation to county governments, accelerating the trend towards greater devolvement of power and spending to the local level. Overall, Odinga’s agenda is likely to be more costly and difficult to implement within the constraints of Kenya’s ongoing IMF-backed fiscal consolidation programme, while Ruto has explicitly pledged to accommodate his promises within existing budgetary targets and resources.

On the anti-corruption front, both candidates' promises ring hollow, in our view. Both are deeply entrenched in the current political system, and are more likely to promote the status quo than pursue radical reforms to structural problems like corruption. That said, Odinga’s appointment of staunch anti-corruption advocate Martha Karua as his running mate gives him the edge on this front over Ruto, whose running mate Rigathi Gachagua is currently facing allegations of misappropriation of funds.

Indeed, a perception of both candidates as tainted old-guard candidates has seemingly contributed to voter apathy, with the participation rate falling starkly to 65% versus 78% in 2017 and 86% in 2013 (although low participation was likely also driven by a severe drought in the north of the country). Further, only 50% of eligible first-time voters registered ahead of elections, meaning the participation rate would have been even lower if the IEBC had come closer to its target for new registrations (this may have disadvantaged Ruto, whose support base skews younger than Odinga’s).

Fixed income implications

For fixed income investors, the key issue heading into the elections was Odinga’s promise to restructure Kenya’s debt. The implications of this statement are still not clear (ie does he plan to restructure all commercial debt, reschedule payments on a bilateral basis with key creditors like China and the CDB, or simply conduct a voluntary NPV-neutral liability management exercise?), but the ongoing eurobond rally despite Odinga’s lead in the pre-election polls (and now the provisional vote count) suggests markets don’t think he would follow through.

That said, under an Odinga presidency the risk of restructuring would loom until his intentions were made clear and continue to weigh on Kenya’s eurobonds, and markets may be a bit too sanguine on this front. Ruto, on the other hand, has said that “I am not going anywhere near or even having a discussion on restructuring debt” and referred to Odinga’s remarks as a “very condescending and reckless statement.” As such, Ruto’s victory would remove one of the key headwinds for Kenyan eurobonds.

Likewise, the risk of post-election violence has likely weighed on bonds in recent months. While the consensus was for limited violence, markets (and ordinary Kenyans) will only be able to breathe a sigh of relief if, and when, the losing candidate concedes defeat. This should boost Kenyan bonds, but if post-election violence does break out then bonds will suffer. And, if the results are challenged in court or if a run-off is triggered, there may still be some time before this uncertainty dissipates (eg over 2 weeks if the results are challenged, 10 weeks if they are overturned, or 6 weeks if there is a run-off).

Beyond these key risk factors, the outlook for Kenya’s economy (and therefore its bonds) will hinge in large part on the new administration recommitting itself to Kenya’s ongoing IMF programme, which runs through May 2024. Neither candidate has voiced opposition to the programme, but given Kenya’s incredibly tenuous balance of payments backdrop (see here for a detailed analysis) there is little margin for slippage, even after two years of outperforming programme targets.

Regardless, many people seem to think that the policy outlook will improve after elections regardless of who wins as pre-election spending pressures dissipate and the new administration leverages its fresh mandate to push ahead with economic reforms. Indeed, a recent survey by the Kenya Private Sector Alliance and Tifa Research showed that most Kenyan CEOs expect economic conditions to improve after the election, and renewed confidence could help revive sagging private sector investment.

While we don’t necessarily disagree with that statement, we think bonds will perform better if Ruto wins due to less anxiety over restructuring and a slightly more market-friendly policy agenda. But either way, contingent on the new administration’s commitment to Kenya’s IMF programme and, if Odinga wins, clarification that he does not plan to restructure external debt held by private creditors (which is a big if), we would tend to agree that conditions should improve with elections in the rear-view.

As such, despite the recent rally, we think Kenyan eurobonds will continue to outperform if and when some of the previously mentioned headwinds are removed, with the ’32 already rallying by 15pts since mid-July to US$77.75 as of cob yesterday. While yields on the ‘32s have fallen below 12% from 15.5% in mid-July, but are still roughly double where they were a year ago and nearly 200bps above the 10% they traded at before the June/July sell-off. Likewise, the yield on the ‘24s has fallen from nearly 22% in mid-July to just above 12%, but still well above the 3% seen this time last year and 10% at the end of May.

Overall, we think there is still upside for Kenyan eurobonds and retain our Buy recommendation across the curve at a price of US$77.75 (11.9% YTM) for the KENINT 8 05/22/2032s as of cob on 10 August on Bloomberg, although that upside has diminished greatly on the back of the recent rally.

Equity implications

The Kenyan equity market arguably already reflects the macroeconomic and political risks ahead and deserves at least its full benchmark weight for FM dedicated funds (c4%). The problem for mainstream EM investors is that the most compelling value is not found in the most liquidly traded stock, Safaricom, but, instead, in the relatively less liquid banks.

Kenya equities, measured by the Nairobi All Share (NSEASI), are down 14% ytd, ahead of Africa peers like Egypt and Ghana (both down about 30%), and much worse than commodity exporter South Africa (down 8%). Notional outperformers Nigeria (up 23%) and Zimbabwe (up 47%) are not comparable because of capital repatriation restrictions. Kenya trailing PB is 1.3x (for 18% ROE), a 23% discount to the 5-year median – similar to the 25% discount to that in Egypt.

Safaricom (57% and 65% of the Nairobi All Share and MSCI Kenya indices, respectively) continues to be virtually the only stock in Sub-Saharan Africa ex-South Africa that is of interest to global EM funds, with just under US$2mn of average daily traded value over the past six months. But that is a curse in an environment of de-risking off benchmark exposure and pressure on EM fund outflows. Safaricom shares are down 21% ytd in US$ total return terms. Yet valuation, with consensus forward PE at 16.4x, is in line with the five-year median. However, that multiple sits alongside negative 10% consensus earnings growth for the current year, compared with growth in the range of 20-30% in the early part of that five-year history, and 8% for the following year – a 4.8% forward dividend yield offers some offset, but not for an investment thesis normally built on growth.

For dedicated FM or Africa ex-South Africa funds able to stomach much lower liquidity (nearer US$0.5mn ADV), the banks look more attractively valued. Equity Bank (8.8% of NSEASI, 24% of MSCI Kenya) is on forward PE of 4.0x, a 10% discount to the five-year median, and forward dividend yield of 7.6%. And KCB (6% of NSEASI, 0% of MSCI Kenya) is on forward PE of 3.5x, a 10% discount to the five-year median, and dividend yield of 8.7%.

Research

Access & Events

Country Risk Index

Sponsored Research

Macro & Market Data