South Africa

South Africa

Interview: Pali Lehohla on statistics, soccer and poetry

Pali Lehohla, South Africa’s statistician-general, shares his thoughts with RYLAND FISHER on mistakes made during the previous census, subjects children should be taught at school, and what he does in his (little) spare time.

Pali Lehohla famously wore only yellow suits during South Africa’s last census, which was conducted in 2011 and concluded in 2012. Yellow is the colour of Statistics South Africa, the organisation behind the Census, which he heads up.

However, dressed in a brown tweed suit last week, Lehohla said the possibility of his yellow suits making a comeback is limited. The reason is that Lehohla is due to retire in 2015, and the earliest there might be another Census is in 2016, with the likelihood that it will only be in 2021.

Interviewed at the Wallenberg Research Centre in Stellenbosch, where he was attending a conference, Lehohla spoke about his passion for statistics, his successes at Stats SA since he took over in 2000, and his plans for the future.

He also spoke about the hurt he felt when he had to let most of his 150 000 staff go after the census, and the pain he felt when his favourite soccer team lost their league title this year. He also spoke about why he encourages young people to study mathematics, statistics and poetry.

When I spoke to a senior person at the Independent Electoral Commission recently, he told me people have the perception that the IEC only works during elections. I suppose people could have a similar perception about Stats SA: that you only work at census time. In a nutshell, what are the main responsibilities and activities of Stats SA?

It brings knowledge to South Africa. It creates the possibility for people to understand what is happening in the nation. This is done through a number of channels. The most conspicuous and the one that stretches the organisation to its limits is the Census. The Census generates population and social statistics.

Then there is a whole series of economic statistics, such as the Consumer Price Index (CPI), which the Governor of the Reserve Bank uses to set rates. We produced more than 220 sets of statistics every year.

We have the privilege of intruding in all homes across the country, and bringing those conversations to the public.

There were some questions after the last census about its integrity. Have those problems been sorted out?

Issues of quality never disappear. The questions remain and continue to be discussed. However, I think we have answered to the best of our ability all the questions that we raised.

The question that surely is a moot question is: what can you do when you have had a 14% undercount? It is too high. How does South Africa reduce the undercount? It is not something that can be done by Stats SA.


Who has to do that?

The public has to address that question.

And that is by participating in the census?

Once the population says it is ready to participate, you have to create conditions to make this possible. I don’t think the conditions for participation are optimal. You have to understand our history, and that it is not very easy to count people who are so mobile.

In other countries, they send census forms by mail. It will be filled in and returned. You can’t do that here. The choice we have made is not to send by address because we don’t know the address.

You have to go there physically. This is partly because of the way residences were designed under apartheid, which makes it difficult to find people at home. But even where there was no apartheid, getting people at home is becoming increasingly difficult because of lifestyles.

There are countries where they say you cannot move until you have been counted. Countries such as Nigeria and Turkey set aside two days for counting. That’s one way in which we can reduce the undercount.

The other way is by using teachers, because they are evenly distributed and they impact on families everywhere. But in the current environment, we can’t entrust that task to teachers because they will fail this mandate. Also, there is a lot of pressure and demands on teachers to teach.

The last undercount was 14% and the one before that 17%. What kind of figure would you be happy with?

Anything in single digits would be good; anything less than 10. If we get five, we would be elated. The whole organisation would be promoted if we got that (laughs). I think we should aspire for 5% but the conditions in the country do not make it possible.

What are the major challenges that you face with making sure that your information is beyond question?

The first is that the public has to be ready to know and answer the questions. That is the area we were able to address quite nicely during the last census. Our campaign was quite powerful. That campaign is continuing, because in the way that we are delivering the results, we are in the public space all the time by making the data accessible.

The second issue is that we also need to target a level of children studying, understanding and using the data as the basis for learning. We can assist them to do so when they know how to count. Once they know how to count, they can make their observations a lot better.

We should get teachers to teach statistics. This will have an almost immediate impact because, as the children grow up, they will be numerate and will deal with numbers in a different way.

We need to see more statisticians in industry and not only in universities. Industry is where you will make change. At universities you only philosophise about life.

Our task as state outside academia is really to change lives, to make a difference. You need a high level of skills in statistics to see what causes change otherwise you can make serious mistakes. Statistical mistakes can be very destructive because they can cause a whole nation to perish.

The small mistake that we had on the CPI in 2003 exaggerated the cost of living and created a lot of suffering in the country.

There are challenges at a middle management level where statisticians get absorbed in their careers as if it is an end in itself and that creates problems. They don’t see the world outside and they don’t see the volume and the value of the work that they do. There is this inward-looking process which is caused by the nature of the work. They have very little sense of what is happening outside.

The area of data mining is an interesting one. By its very name, data mining means that you have to go so deep into the data that you can actually be so absorbed and forget what its use is.

The other challenge is the accountability of statistics offices. In this environment there is a very strong likelihood of political interference. We have seen this in countries such as Argentina, Greece and Italy.

In South Africa I don’t expect that because we respect the law. The law is very clear and the practice for the last 19 years has been very clear that there shall not be any attempt at political interference in the production of statistics.

We’ve worked with Trevor Manuel, who has been the minister responsible for statistics for more than 17 years. He has entrenched a principle where there is the law but there is also the practice. The law and the practice have been coherent.

However, it is important for the leader of statisticians to know how to interact with politicians.

You have done lots of work internationally. How does South Africa fare in relation to other countries when it comes to the collection and dissemination of information about its population?

South Africans unfortunately don’t like themselves, that’s what I generally see. I don’t say it lightly and I don’t know how else to say it. That statement has two contradicting positions and I don’t know what brings it about. If you look at the leadership of South Africa on the continent and the leadership of South Africa elsewhere, we really have such enormous social capital to really get our heads together and focus. In a way, the National Development Plan should help us to turn our own self-deprecating into a different way of looking at things.

In relation to statistics, we are actually a leader in that, the way we have actually upped the ante on dissemination. There is no country that has done this.

You have embraced quite a lot of technology?

Yes, we are there. In terms of being daunting, of trying new things, we are there. In terms of the quality of the data that we have produced, we are probably one of the top 10 or 15 in the world.

We are doing very well. I’ve chaired the United Nations Statistics Commission and we are in a good space.

On the continent, we have handled the symposium on statistical development and we have changed the way statistics are collected as a result of that.

For all of this, Manuel has been either in the lead or on the side or at the back or underneath. He has really steered us well in terms of opening up opportunities.

We tend not to recognise the pearls that we have in the country. That sometimes leads to us losing opportunities.

What kind of staff count do you have at Stats SA and how much do you bump it when your activities are at a peak?

We have 3 800 and, at the time of the census, we have the equivalent of 10% of the civil service, at about 150 000 people. The civil service is about 1.5-million people. Imagine an organisation with about 3 000 people managing 10% of the civil service. It’s quite a challenge to step up in this way. It’s a peak that is difficult to manage. We have to recruit, train, employ, pay and decommission.

How do you ensure the staff you employ, especially the part-time ones, deliver on your mandate?

Our work ethic is what drives us. It is always painful when you know that workers are temporary and that they are going to go. There is not much you can do about it, but the camaraderie of feeling their pain beyond lip service, is what makes them do their best and do the right thing.

When we have tight timelines, discipline is very important and where discipline is violated, you have to be very ruthless. For instance, there was a situation where guys wanted to disappear with our questionnaires and we searched for them the whole night. By the morning we got them to count the questionnaires and deliver them.

We need to be extremely ruthless, not in a punitive manner, but fair. We need to understand the important work that these people do.

You also have to follow up on everything that comes to your phone and that is what I do all the time. I hardly slept during the census period. I got calls about people who did not get paid and had to ask someone to drive there to make sure people got paid.

This is what you have to do. You have to make sure that you do everything to ensure that these people get the justice they deserve.

Do you get a sense that decision-makers, particularly in government, use the information you provide them or is there room for better use of your information?

At a macro level, I would argue they use it, if you look at how rates are set, how the budgets are set, how they are allocated. At the level of execution, implementation, using the money, I don’t think people use the data that much.

When one looks at urban, regional and spatial planning, that’s when you see the disjuncture between the intention, the resources and the actual execution.

In large part, there is still a lot of ignorance. There is a lot of bad quality data that is being used.

You have been described as a man who lives and breathes statistics. Have you always known that this is what you want to do with your life?

Not at all, but there was a turning point. When I was growing up in Lesotho, I did not really know this was what I wanted to do, except in the naming of an ox who we called census way back in 1965, when I heard this term for the first time.

How old were you then?

I was born in 1957 so I was about seven-and-a-half. My father had brought a brown ox from the Free State. Afrikaner farmers don’t give names to their livestock, so when this thing arrived, we had to give it a name. We called it census because there was going to be a census in Lesotho.

There were three turning points in my life. The first one was when I arrived at university and I did not know what to study. I wanted to do geology, but there was no geology in Lesotho. I ended up doing statistics, with economics and political economy. It was a strange combination but I found it very interesting.

The second turning point was when I met Professor Manie Geyer in 1993. He made me understand how space could be used in a democracy, based on statistics.

The third turning point also came when I was at university and I met a student who told me about how he only wanted to be a revolutionary.

I laughed when he said that, but when I look at how statistics, I see it has influenced the way I think and the way I see the need for change. It’s almost a constant desire to change things and influence change in a positive way.

You are known to have consistently worn yellow suits during the last census. Have we seen the last of those suits or will they make a comeback for the next census?

Well, I have two-and-a-half years to go as the statistician-general and the question of a census is still an open one. The [Statistics] Council was asked by Manuel to look into the matter as to whether we should have a census every five or 10 years. That matter has not been resolved yet, but the preference appears to be for a 10-year interval.

There are two things working against the yellow suit. The first thing is that by the next census in 2016, I would have thrown in the towel. The second thing is that there might not be a census. And I’m not sure whether it was not a once-off process. I don’t think one can put it on again and think you can win. I think one has to do something different, but it was good while it lasted.

You have been at the helm of Stats SA since November 2000, but you are still relatively young at 55 (he will be 56 in September). How much longer do you see yourself in this position and what else would you do?

When you run institutions, you have to think about exits, when it is appropriate and proper. I would have spent 15 years in the organisation by that time.

The platform for production of statistics in the organisation is good. I don’t think we can make silly mistakes like we did in 2003 or 2005.

What I have to worry about is whether the people who produce that platform continuously get better at it. We must have continuous improvement through better staffing policies, staffing processes and so on.

If I were to continue working, I would not see myself as the person supervising Stats SA. I would see myself as a person who campaigns for change in the use of evidence and ensuring that statistics is taught in schools.

The reports, the data that Statistics SA produces would not be my worry at all. I see myself having a very small responsibility for the production of data.

If there is a Census in 2016, I don’t see myself really running it, unless there is a crisis, because when a crisis comes, you just need to go in there and do that work. I see myself playing more of a supporting role.

I believe that 2013 is the Year of Statistics. What have you done, and what are you still planning to do, to mark this occasion?

I am the vice-president of the International Statistics Institute (ISI), the parent body of the Year of Statistics, which UN secretary-general Ban Ki-moon launched during the United Nations Statistical Commission in New York. The responsibility for promoting the Year of Statistics rests largely with me on the continent and in South Africa.

This process we launched this week, the ISIbalo Symposium on evidence-based decision-making, is part and parcel of expressing our support and self-interest in the Year of Statistics.

There are several other activities planned by Stats SA. We are co-hosting a major conference in Johannesburg in October, we will be having a lecture series on statistics, and we are looking at launching a book on statistics towards the end of the year.

Of course, there is the World Statistics Congress in Hong Kong in August, which is following on the International Statistical Institute meeting, which was hosted in Durban in 2009.

From where you sit as a statistician, what can you see as the biggest problems facing South Africa ahead of us celebrating 20 years of democracy?

What we have to celebrate is the speed with which measurement was put at the centre of the agenda. What we have to lament, though, is the absence of use of measurement that emanated from this.

There are many missed opportunities in the absence of the observation of data and the use of data.

The practice of measurement is not only the role of Stats SA, but of everybody else. That agenda still has to be driven very, very hard. I value that in the 20 years of democracy, you have seen changes in institutions, and an entrenchment of the culture of measurement.

The next phase is that the practice of measurement now has to grow roots both in the state institutions and in the teaching of children, so that the next 20 years will be very different and opportunities will be understood, taken and used when they arise.

We cannot lament what missed opportunities we had.

I think we took too long to get something like the National Development Plan on the table after the RDP (Reconstruction and Development Plan) was shelved.

It’s important to reflect on these things, because we don’t confront the evidence and consequences of what we did.

What kind of advice would you give young people who are considering a career in statistics?

Whether they want to undertake a career in statistics or in any other field, I would argue that they need to sharpen their mathematics skills. Not all people will study mathematics or pass it, but everyone should do it up to a particular level. It actually helps you to clarify certain things.

My encouragement would be: do mathematics, do statistics because it is a service subject for everything you do, even medicine. You should also do poetry in whatever language you choose. It refreshes your mind and gives you a different dimension.

When you bring these three things together, you realise that you understand the world in a much better way.

What do you do with your spare time, if you have any?

My wife usually complains about that. There has not been any. My work causes me to travel. When I am on the plane, I have time to think and when I am away I have time to think and write ideas. When I come back I push them back in the organisation.

My biggest problem is not knowing how to relax. Of course, I go to funerals and weddings and that’s good enough (laughs).

And you don’t watch sport?

I do. I do, but the problem with watching sport is that you get attached. There is a club that I used to watch regularly with my youngest son, because we like the same club, but now, the pain of losing is so bad that I don’t even watch the soccer any more (laughs). I am not going to mention which club it is. I am hopeful that it will bounce back and I will be watching again. DM

Photo: Statistician-General Pali Lehohla releases South Africa’s gross domestic product (GDP) for the first quarter at a news conference in Johannesburg on Tuesday, 28 May 2013. Picture: Werner Beukes/SAPA