Rwanda’s Kagame on the Social Construction of Ethnicity

This is from an interesting interview with the FT:

During the interview, Mr Kagame says it matters little whether there are real physical differences between Hutus and Tutsis or whether these were arbitrary distinctions codified by race-obsessed imperialists. “We are trying to reconcile our society and talk people out of this nonsense of division,” he says. “Some are short, others are tall, others are thin, others are stocky. But we are all human beings. Can we not live together and happily within one border?” Mr Kagame has taken a DNA test that, he says, reveals him to be of particularly complex genetic mix. The implication, he says, is that he, the ultimate symbol of Tutsi authority, has some Hutu in his genetic make-up.

The transcript is available here. Read the whole thing.

Also, the average Rwandese lives a full six years longer than the average African.

Screen Shot 2017-08-28 at 11.13.48 AM.png

Ultimately, the sustainability of Kagame’s achievements will depend on his ability to solve an important optimal stopping problem:

The problem, he says of who might succeed him, is preventing someone from “bringing down what we have built”. Above all, he says, he wants to “avoid leaving behind a mess”.

The president insists it was never his intention to stay on, but the party and population insisted. “We are not saying, ‘We want you forever until you drop dead,’” he says, imitating the voice of the people. “We’re only saying, ‘Give us more time.’”


On NYT’s Misguided Nostalgia for Conrad

The smoked monkeys brought the point home. During my first day on a boat on the Congo River, I’d embraced the unfamiliar: how to bend under the rail to fill my wash bucket from the river, where to step around the tethered goat in the dark and the best way to prepare a pot of grubs. But when I saw the monkeys impaled on stakes, skulls picked clean of brains and teeth thrusting out, I looked otherness in the face — and saw myself mirrored back.

I was the real exotica: the only tourist to take this boat in nearly a decade, and the only white woman, as far as the crew knew, ever. Expect to be kidnapped, people had warned me. Expect to have everything stolen and expect every arrangement to go awry. Bring your own mosquito net, waterproof everything twice and strap your cash around your ankle.

These are the opening paragraphs of Harvard historian Maya Jasanoff‘s piece on the Congo River in the Times. In the piece Prof. Josanoff seeks to uncover what has changed (or not) since Conrad was in the Congo a century ago. She has a forthcoming book titled The Dawn Watch: Joseph Conrad in a Global World.

The piece reads like a satirical rejoinder to Binyavanga Wainaina’s How to Write About Africa.

Why did Jasanoff decide to embark on this mission? In the third paragraph we find out:

The Democratic Republic of Congo, I read in my guidebook, was “a huge area of dark corners, both geographically and mentally,” where “man has fought continuously against his own demons and the elements of nature at large.” This, in other words, was the heart of darkness, which was why I had wanted to come.

I would like to think that Prof. Jasanoff consulted more than a guidebook to find out more about the Congo before her trip. But I digress.

The whole thing is worth reading. After which you should write the New York Times editors to let them know what you think.

The point here is not necessarily to call out Prof. Jasanoff, but to highlight what seems to be an insatiable demand at the Times for orientalist pieces on Africa and Africans.

And while on the subject of Conrad and the Congo, a fitting antidote is by none other than Chinua Achebe. In his review of Heart of Darkness, Achebe laments the use of Africa and Africans as background against which Europeans act out their neuroses:

Africa as setting and backdrop which eliminates the African as human factor. Africa as a metaphysical battlefield devoid of all recognizable humanity, into which the wandering European enters at his peril. Can nobody see the preposterous and perverse arrogance in thus reducing Africa to the role of props for the break-up of one petty European mind? But that is not even the point. The real question is the dehumanization of Africa and Africans which this age-long attitude has fostered and continues to foster in the world. And the question is whether a novel which celebrates this dehumanization, which depersonalizes a portion of the human race, can be called a great work of art. My answer is: No, it cannot.

Read the whole review here.

Initial Thoughts on the Reelection of President Uhuru Kenyatta

Congratulations to President Uhuru Kenyatta on reelection. In the end, he outperformed the polls by having a well-oiled national campaign that paid close attention to down ballot races. Jubilee MPs, Governors, Senators, and MCAs were elected in Bungoma, Kakamega, and other key swing areas. Where the party won it won big; and in the places it lost, it stayed competitive. The same cannot be said for NASA-affiliates.kenyatta

The polls were not that off. Kenyatta led in all but one poll conducted by Infotrak. In the end it appears that the undecideds stayed home. Turnout was relatively lower in Western and Coast regions (to regions with the biggest share of undecideds) relative to the national average. Odinga needed to at least match Kenyatta’s stronghold turnout in these regions to stay competitive.

Of the two models that I ran, the one incorporation registration rate as a measure of voter enthusiasm did better that the one that only considered historical turnout rates by region. Kenyatta supporters registered at high rates and followed through on Tuesday. Undecided Odinga supporters stayed home on Tuesday.

In a model that gives less weight to registration rates (as proxies for voter enthusiasm and likely turnout), the estimated vote share is Kenyatta 52.8% vs Odinga 47.2%. 

A more involved model that tries to estimate differential voter enthusiasm yields an even bigger advantage for Kenyatta (54% vs 46%). 

Turnout in 2013 was most certainly inflated by both CORD and the Jubilee Alliance.

Whatever one thinks of him, William Ruto is a political genius. After Tuesday has emerged as arguably the most powerful politician in Kenya. Initially I had thought that Jubilee Party was a bad idea that would end up depressing turnout by forcing everyone to vote for the same candidates. In the end it did not matter. Instead, Jubilee won big in the presidential election and, perhaps more importantly, swept key down ballot races. The party will command at least 49% of the seats in the 12 Parliament and will most certainly hit more than 50% with the support of friendly independents. Ruto has successfully vanquished the Moi family in Rift Valley politics. And more importantly, he is slowly emerging to be a national politician with a strong direct following outside of his core base. Only Odinga has managed to achieve this feat in the recent past.

Chances are very high that William Ruto will be the 5th President of Kenya. I must admit to have been wrong in assuming that his political stock would plummet as soon as Kenyatta won reelection. Instead, I think because of his hold on Jubilee his stock will only rise with time. Kenyatta’s elite base cannot push him aside. He has the numbers in Parliament and the very credible threat of inflicting maximum pain by raising political temperatures in the Rift Valley.

This will be a tough loss for Odinga supporters. At 72, this was surely his last stab at the presidency. There will be a lot to be said about the organization and strategy of his campaign — including the apparent lack of polling agents, failure to try and raid Jubilee strongholds, and own goal regarding the prospects of violence following a rigged election (the latter may have cost Evans Kidero the Nairobi governorship). There is also the issue of IEBC’s inability to relay results with the confidence of all parties concerned. However, despite the possibility of hacking of the results transmission system, the down ballot results point to a credible Kenyatta win. Unless more evidence becomes public, I am inclined to believe that this election was credible. The KIEMS system worked. IEBC should build on this success to strengthen the transmission system. It was messy, yes. But it was also most certainly better than last time.

This is a step forward in Kenya’s political development. The opposition is in disarray, but the real institutional fights are about to start within the Jubilee coalition. Ruto’s 2022 ambitions will likely force him to expand the size of the Jubilee coalition. He will likely reach out to Odinga’s base — including in Coast, Nyanza, Eastern, and Western regions. Kenyatta will also want to give himself some credible check on Ruto’s power and influence, which will force him to reach outside of Jubilee’s core constituency for institutional support. Recall that Jubilee is now pretty much Ruto’s party. It will be interesting to see how Joho, Nyong’o, Ngilu, and Oparanya react to all this.

I expect Kenyatta to be constrained by intra-Jubilee politics in his second term. I will say more on the likely political and policy direction of the second term after Kenyatta announces his new cabinet.

In the next fortnight I’ll probably put together a piece on the historical and political significance of Odinga’s likely exit from the political stage.

Hon. Raila Amolo Odinga is perhaps the one individual who has contributed  the most to democratic consolidation in Kenya since the early 1990s. He is also a tragic figure who has had to deal with personal shortcomings, family tragedy, and systemic rejection by Kenya’s powers that be — all played out in full view of the Kenyan public. In addition, it is impossible to talk about Odinga without mentioning the ethnic factor. I think that the biggest impact of this loss will be the Kenyanization of the Luo elite. Since Odinga Senior, the Luo elite have invested a lot in trying to change Kenya (at great expense for the Luo masses) — a fact that made Luo Nyanza the perennial epicenter of oppositionist politics. But with Odinga’s exit, this collective commitment to oppositionist politics will likely diminish. I expect Luo Nyanza politics to become more fragmented and transactional (i.e. less purist). All else considered, the Kenyanization of the Luo elite will probably be a good thing for the masses in Luo Nyanza.


Three Known Unknowns Ahead of Kenya’s Presidential Election Tomorrow

Here are three factors that could potential lead to a surprise outcome in Kenya’s presidential election tomorrow.

1. Turnout: To be honest, we don’t really know what to expect with regard to turnout figures tomorrow. As Charles Hornsby notes here:

For some time I have been wrestling with an ethical problem. Reviewing the 2013 turnouts, in comparison with that from previous national elections since 2002, it became clear with the benefit of hindsight that turnouts were implausibly high not just in Luo Nyanza and Central Province, but in many other places…….. The size and scale gap between 2013 and every other election for the past 15 years is hard to explain.

In other words, there is good reason to believe that the 2013 turnout figures were artificially high.

The use of technology and a raft of reforms, including the planned announcement of results at the constituency-level, are supposed to minimize opportunities for political parties to gin up turnout in friendly areas this time round. If these safeguards work, we will all be flying blind with regard to our turnout-based predictions of the final outcome in the presidential election. All I can say for now is that Kenyatta has a structural advantage over Odinga on likely turnout, but that if Odinga matches Kenyatta on turnout (especially in Coast and Western regions) he will most likely become Kenya’s fifth president.

2. Technology: The Kenyan public’s perception of the integrity of the polls will crucially hinge on how well the IEBC’s Kenya Integrated Electoral Management System (KIEMS) is judged to work. Kenya is holding one of the most expensive elections in the world (judged by cost per vote). And the use of expensive technology has been touted as a way of minimizing human discretion and, therefore, opportunities for vote rigging. But technology only works as well as humans allow it to. And sometimes stuff hits the fan. In addition, the IEBC only partially tested its results transmission system. Which means that there will most likely be hiccups in the data transmission process from the country’s 40,883 polling stations. Widespread technological failures will tarnish the integrity of the outcome, and could be a catalyst for political instability. 

3. The accuracy of opinion polls: Throughout this cycle, all polls except one have shown Kenyatta to be ahead of Odinga. In the recent past the polls have certainly tightened, with Kenyatta leading Odinga by 3 percentage points or less. But what if the polls are wrong? In 2013 they underestimated Kenyatta’s support by about 2 percentage points, on average. The same inherent bias may be at play this year, or it could be reversed. One important factor to look out for will be the effect of local gubernatorial races on the presidential race. For instance, the outcomes of local contests in counties like Machakos, Bomet, Narok, Meru, and Bungoma will likely have non-trivial effects on the presidential race in the same counties. Yet throughout this cycle there has been very little effort to calibrate the national polls using information from the state of county contests. And so while I believe that existing opinion polls give a fairly accurate depiction of the state of the race, only the final vote counts will tell.





What the poll numbers tell us about Kenya’s presidential election next Tuesday

Below is a table from a previous post comparing the poll numbers and actual votes ahead of the 2013 presidential election, as well as May 2017 poll numbers for the two leading candidates (Messrs. Uhuru Kenyatta and Raila Odinga). You can read the background post here. See also here.Screen Shot 2017-05-31 at 9.30.40 PMAlso below is the July 23rd poll released by Ipsos Synovate that reveals some interesting changes over the last two months. As of now the race stands at 47% vs 43%, advantage Kenyatta. But with a margin of error of 2.09%, this is a statistical dead heat.

On President Kenyatta: 

Kenyatta has seen his support decline in Nairobi (by 11 percentage points), Coast, Eastern, and Western regions. His support has increased in North Eastern (by 5 percentage points), Nyanza and Rift Valley regions. In addition, his support is stable in Central region at 88%.

On Prime Minister Odinga:

Odinga’s support has declined in Rift Valley (by 6 percentage points) and Western. His support has increased in Nairobi (by 13 percentage points), Eastern, Central, Coast, and North Eastern regions. In addition, his support is stable in Nyanza region at 76%.

Screen Shot 2017-08-02 at 8.08.46 AM.png

Looking at Trends:

A bigger share of undecideds will likely break for Odinga: The above trends confirm my previous observation that undecideds in Coast, Eastern, and Western regions are likely to be reluctant Odinga supporters. Odinga has seen his poll numbers go up in Coast and Eastern as the number of undecideds has shrunk in both regions. At 14% in the latest poll, the number of undecideds in Western region remain virtually the same from the May figure (16% of respondents). Both Kenyatta and Odinga saw declines in their support in Western region between May and July, but most of the undecideds in the region will likely break for Odinga. Virtually all the leading political elites in the region support Odinga’s bid for the presidency.

The race has tightened over the last two months: It is also clear that the race has tightened over the last two months as more voters have internalized the fact that this is a two-horse race between Kenyatta and Odinga. In particular, much of the tightening appears to have come from shifts in Western and lower Eastern, the two regions comprising the “Big Five” voting blocs where a clear majority of elites are behind the Odinga ticket.

Screen Shot 2017-08-02 at 8.03.45 AMScreen Shot 2017-08-02 at 7.59.37 AM

What do the numbers say?

First a couple of caveats.

In the last cycle the polling by Ipsos was off by a few percentage points in either direction. The final polls before the election underestimated Kenyatta’s support in all regions except Western. The final poll also underestimated Odinga’s support in Coast, Nyanza, and Western regions, was spot on in the Rift Valley, and overestimated his support in the remaining four regions. This is largely because polling in Kenya is often structured by region, on account of the fact that vote choice often maps neatly on regional/ethnic cleavages. No polling firm has figured out a likely voter model (I honestly don’t know why). Much of the underestimation of Kenyatta’s support came from the erroneous assumption of evenly spread turnout across the country.

And speaking of turnout, it is important to note that the Uhuru/Ruto ticket has a structural advantage. On average, Kenyatta’s core support comes from wealthier regions of the country that are relatively easier to reach and mobilize. In addition, with nearly 40% of the electorate coming from just two of the “Big Five” voting blocs, the combined ticket also offers a classic minimum winning coalition which adds to efficiency of messaging and turnout mobilization. This is in contrast to the Odinga/Kalonzo ticket’s turnout challenge. Its core is three of the “Big Five” that combined add up to just over 30% of the electorate. Therefore, Odinga has to make the difference by appealing to smaller voting blocs, particularly in the Coast, North Eastern, and Rift Valley regions (especially among pastoralist subregions of the Rift). The pro-Odinga non-Big Five regions have historically had relatively lower turnout rates, in no small part because of lower rates of access to education and economic opportunities. Dispersed support also means dispersed messaging. It is not clear that Odinga has successfully been able to overcome his turnout challenge ahead of Tuesday’s election.

The numbers, taking into account likely turnout: 

Based on 2013 turnout, I have created a model that takes into account county-level turnout (averaged at the constituency level to generate more data points). I have then clustered individual counties into regions. A more granular model that looks at individual voting blocs (e.g. that separates Nyamira and Kisii from the rest of Nyanza, Kitui, Makueni, and Machakos from the rest of Eastern, or Narok from the rest of the Rift Valley yields more or less similar results). Lastly, I weighted the estimated turnout in each county by the registration rates ahead of the 2017 elections. Turnout in 2013 varied from a low of 58% in Kilifi to a high of 95% in Makueni. Similarly, registration rates (as a share of eligible adults) ahead of 2017 varied from a low of 50.6% in Vihiga to a high of 86% in Kajiado. Presumably, these differences in registration rates reveal information on voter enthusiasm, and therefore likelihood of turning out next Tuesday.

All this is like attempting to perform open heart surgery with a blunt panga. So bear with me.

After the weighting, I then estimated turnout rates at the county level after which I used the raw vote numbers to estimate turnout rates at regional levels (polling is sparse at the county level). With the regional turnout figures, I then estimated the likely vote totals for Kenyatta and Odinga by region. Throughout this process I ignored the proportion of voters that are undecided.

In a model that gives less weight to registration rates (as proxies for voter enthusiasm and likely turnout), the estimated vote share is Kenyatta 52.8% vs Odinga 47.2%. 

A more involved model that tries to estimate differential voter enthusiasm yields an even bigger advantage for Kenyatta (54% vs 46%). 

In short, going by historical turnout rates, Kenyatta is still a strong favorite to win reelection next Tuesday.

Now, there are several ways in which I could be totally off the mark.

First, there is the issue of undecideds. The bulk of these voters are in Coast and Western regions, both Odinga strongholds. So far the trends indicate that undecideds appear to be breaking for Odinga in larger proportions. Should this trend continue, Odinga may eat enough into Kenyatta’s leads to force a runoff, or even an outright squeaker of a first round win.

Second, there is the issue of turnout. If Odinga were to average a turnout rate of 85 in Nairobi, Coast, and Western regions, my model estimates a vote distribution of 50.9% vs 49.1% in favor of Odinga. If Western and Coast regions alone got to 87% turnout, Odinga’s lead would increase to 51.1% vs Kenyatta’s 48.9%. Structurally, Kenyatta pretty much maxed out on turnout in 2013, while Odinga has a lot of head room.

As I keep saying, this is going to be very much a turnout election in which historical voting patterns strongly favor Kenyatta. For Odinga to have a fighting chance he has to convince undecideds to turn up and vote for him next Tuesday.

Other Interesting Polls to Consider:

There are two other polls that came out yesterday, one commissioned by Radio Africa and another done by Infotrak.

The Radio Africa poll puts the race at 47% for Kenyatta vs 46% for Odinga, a virtual tie. With this poll, too, the trends show a tightening race. Earlier in the month Kenyatta led Odinga by 49% vs 44%. The Infotrak poll puts the race at 49% for Odinga vs 48% for Kenyatta. Infotrak is the only polling company that has shown Odinga leading Kenyatta throughout this cycle.

Finally, Ipsos also released a poll yesterday that puts the race at 47% (Kenyatta) vs 44% (Odinga), but the details of which are not yet available. More on this soon.