Johan Fourie's blog

I'd rather be a comma than a fullstop

Archive for November 2015

What determines South Africa’s inflation expectations?

with one comment

Do you know your financial adviser's biases?

Do you know your financial adviser’s biases?

Our perceptions about expected inflation have important consequences. It influences how we (or the trade unions we belong to) negotiate wages, our willingness to buy a house, and businesses’ decisions to invest or not.

Inflation determines how many goods and services we can purchase with our current salary. Inflation expectations, however, determine whether future goods and services will be more or less affordable and therefore how we will behave today. If we believe inflation will be 10% next year and our salary increase only 5% (leading to a decline in our purchasing power), we might be less likely to buy that new car.

Measuring inflation expectations and determining what influences peoples’ inflation expectations is therefore important to policy makers, notably, in South Africa’s case, the SA Reserve Bank (SARB).

Controlling inflation expectations is the first step to controlling inflation. If there is a shock to a particular price (such as oil), the macroeconomic consequences will be limited if South Africans believe that Governor Lesetja Kganyago and his Monetary Policy Committee (MPC) will bring inflation under control. If the response of the rest of the economy to the oil shock is muted, the process of inflation won’t get under way.

What do we know about the determinants of inflation expectations? Not much, it turns out. It’s quite difficult to measure inflation expectations. Surveys are often used, but usually of economists in the financial sector; they don’t necessarily gauge the perceptions of the proverbial man-on-the-street.

A new paper published in Economic Modelling by Stellenbosch University economist Monique Reid begins to shed light on the topic. She uses ten years of data from an inflation expectations survey by the Bureau of Economic Research to investigate how quickly the message from the SARB trickles down to the general public. Information is ‘sticky’, Reid finds, and for some groups more than others: financial analysts, for example, adjust their expectations quicker and more accurately than businesses and labour unions. This is because financial analysts have the ability and skill to use and understand other sources of information (like MPC announcements or international economic indicators) than simply the past inflation rate.

A recent paper by two US economists, Ulrike Malmendier and Stefan Nagel, in the The Quarterly Journal of Economics shows that even amongst the general public there is significant variation in inflation expectations. They find that own life experiences determine how individuals form expectations of the future. How would this work?

Say inflation over the last five years has been lower than the average for the last three decades. They show, using 57 years of data of US inflation expectations, that a young person who entered the job market five years ago would be more likely to expect lower inflation than someone who had been employed for longer and had thus experienced both high and low inflation regimes.

If this is true for SA, young South Africans are more likely to expect lower inflation, because the inflation rate in the sixteen years between 1999 (the year SARB started with inflation-targeting) and 2014 averaged 5.2%. South Africans entering the job market during this period would have experienced a low-inflation regime. But those entering the job market in the sixteen years between 1976 and 1991 witnessed an average inflation rate of 14.1%.

This has important implications for policy makers, financial sector managers and their clients. Behavioural economists repeatedly demonstrate that people have biases that they are often unaware of (in this case, a mix of the availability and familiarity heuristics). Younger financial advisors may weigh the recent low-inflation regime more heavily than older advisors would. An obvious way to diffuse such unknown biases is to have teams with advisors of different ages.

Such biases also influence household borrowing and lending behaviour. Younger people may, for example, be more inclined to choose variable-rate investments given that during their lived experience rates never varied dramatically. Older homeowners who remember the trauma of the 1998-rate hike, may prefer fixed-rate investments.

Policy makers in the SARB need to take heed of these findings and include demographic characteristics of consumers into their models of inflation expectations.

*This article first appeared in the Finweek magazine of 29 October.

Our watershed… opportunity

leave a comment »

SlingShot: Could this device make water infrastructure redundant?

SlingShot: Could this device make water infrastructure redundant?

South Africans are gearing up for a tough summer. While loadshedding (the frequent power shortages that plagued the country since 2008) has eased, a new issue came to the fore this week: watershedding. In areas of Germiston and the East Rand, water shortages have been reported and watershedding between 10am and 3pm is on the cards. While the heatwave and drought across the country are the immediate cause of the shortages, there are far deeper structural problems that is likely to make watershedding a common feature of daily life in South Africa in the near future.

This should not come as a surprise. There has been several warnings in the media about the likelihood of watershedding over the past few years, well before the heatwave and drought made the situation acute. On 25 June, Niki Moore wrote the following in the Daily Maverick:

The reason for any potential water shedding is almost a mirror image of why we have load-shedding. Since 1994, millions of people have been added to the water grid with very little thought being given to increasing the capacity of water storage or water intake plants. Combined with mismanagement of water, non-payment for water, huge water wastage through lack of maintenance and neglect, and poor governance through corruption, we are facing a high noon of water shortages that might start affecting us in as soon as a few months.

Well, that was pretty accurate. But watershedding – a shortage of water – is only one concern. The more serious one is the pollution of existing drinking water, which is likely to have serious health consequences. Here is Anthony Turton on the water crisis:

The possibility of major public health crises in the short to medium term is growing and can no longer be discounted. We could soon see a major bloom of toxic cyanobacteria, especially in the light of the increased water temperatures likely to result from the El Nino Southern Oscillation now evident in southern Africa. The growing risk to both companies and individuals needs to be anticipated and understood, so that remedial action can be taken as quickly and effectively as possible.

The reason this is unlikely to happen, he argues, is political:

All available data suggests there is little in South Africa’s water sector to be optimistic about. The level of politicisation has become so high that decision-making is no longer rooted in hydrological realities. Ideology is regarded as paramount, while reality counts only as a secondary factor. The ideological filters in place make it very difficult to carry out any serious technical assessment of water quality or management. In addition, no serious attempt is currently in place to embark on evidence-based policy reforms.

Perhaps the greatest failure of the new order since 1994 has been deteriorating water quality. This has been caused primarily by massive failures in the management of municipal wastewater treatment plants, which have made the State the biggest polluter of water in the country. This looming disaster could have been avoided by more rational and less ideologically-driven policy choices. We need to challenge this approach if we are to re-invigorate our democracy and extricate ourselves from the horns of the dilemma arising from the politicisation of water in a highly water-constrained national economy.

So what can be done about this? The easy but unsatisfactory answer is that local government elections are next year, and South Africans should use this opportunity to demand change. But much of the problem is more systemic and pervasive than local governments can solve on their own; as Turton argues, the “first essential requirement is a new and technically robust national strategic plan for managing, conserving, and augmenting the country’s limited water supplies”. National elections, though, are only in 2018, and it is difficult to envisage dramatic change.

Instead, South Africans will have to find solutions to the water crisis outside the remit of government. This is difficult with a water utility, which is the epitome of a natural monopoly and the reason the state is almost always involved. But just as the case with loadshedding, technology may provide a solution.

Last night, I watched a 2014 documentary about the SlingShot, a device developed by the creator of the Segway, Dean Kamen, to purify water. Much like Elon Musk has created the PowerWall to allow households access to electricity at all times, the SlingShot is a device which would allow anyone able to convert contaminated and filthy water (or even seawater) into drinkable water. The device has been tested in Ghana, South Africa and in several Latin American countries. It has the support of Bill Clinton and the CEO of Coca-Cola.

I’m a technoptimist. I don’t know whether this particular technology will be successful, but as Estian Calitz and I argued in this 2009 paper, technology will allow public goods to be increasingly viewed as private goods, or natural monopolies to be made into competitive industries. Cell phones are a good example, breaking down the need to have large, fixed-line networks that doesn’t make sense to build more than once: i.e. the natural monopoly of Telkom. Elon Musk’s PowerWall, coupled with renewable energy generation, will break the natural monopoly of Eskom. Perhaps Kamen’s SlingShot (or a similar device) will do the same for water.

South Africans have many reasons to be pessimistic about the future. But, as David Landes writes in the Wealth and Poverty of Nations, it pays to be optimistic.

In this world, the optimists have it, not because they are always right, but because they are positive. Even when they are wrong they are positive, and that is the way of achievement, correction, improvement, and success. Educated, eyes-open optimism pays; pessimism can only offer the empty consolation of being right.

Maybe the watershed moment for South Africa is when we realise that the promise of a better life for all lies not in the next elections, but instead in embracing new technologies.

Written by Johan Fourie

November 12, 2015 at 09:44

More math and science, please

with 2 comments

Princeton University professor Leonard Wantchekon was the keynote at the African Economic History Workshop in Wageningen, the Netherlands

Princeton University professor Leonard Wantchekon was the keynote at the African Economic History Workshop in Wageningen, the Netherlands (31 October 2015)

I had the good fortune last week of meeting Leonard Wantchekon, Professor of Economics at Princeton University. His is an incredible story: student protester in Benin, sent to prison, tortured, escaped, fled to Canada, shifted from math to economics, and moved up the ranks to be professor at an Ivy League university.

But apart from these gripping experiences, he noted how important the study of mathematics was in his life, and still is for many students in Benin, a country ranked 167th poorest out of 187 countries in the world. That is why six of the top ten mathematicians in Africa are from Benin (his fact), and why he has opened the African School of Economics in the country.

This got me thinking about a topic I’ve written on before: what do the best and brightest South African high school matriculants choose to study? Of course, their choice is influenced by a multitude of factors. Parents have certain expectations, friends weight in, teachers have their say and there are often financial realities at play. They see adverts for different occupations in newspapers and online, they watch series which portray romanticized images of certain careers, and they dream about working with people, or with animals, or ‘not-in-an-office’. (If I had a Rand for every time a student told me that they just don’t want to work in an office… and these are Commerce students. Buddy, you’ll plead for an office after you’ve spent a few years in a cubicle!)

But a new paper by Biniam Bedasso of ERSA suggests that there are other factors, too, which shape our behaviour. Your science teacher is one. The most significant determinant of choosing a major at the University of Cape Town, according to Bedasso, is the number of science courses an applicant took in high school. The more science courses you take, the more likely you are to choose high-earning degrees like Engineering. Not all schools, however, are equally endowed with good science teachers, which means that inequalities at school translate into inequalities at university: black students who are more likely to go to schools with no science teachers are more likely to end up choosing degrees in the Arts and Humanities, for example.

Peer pressure is another factor that influences degree choice. Using enrollment at UCT between 2010 and 2013, Bedasso finds that if your friends choose a Humanities degree, you are 10% more likely than someone with your exact same characteristics that live elsewhere to also choose a Humanities degree. In his words:

Neighbourhood effects shape the choice of individuals through the influence of near-peer role models. Correcting for possible clustering of unobserved preferences along postcodes, a one standard deviation increase in the ratio of near-peers who were admitted to a certain faculty during the last three years is shown to increase the probability of choosing the same faculty by around 10 percent.

He also finds, interestingly, that politics matter. Black South African matriculants are more likely to choose Commerce or Arts degrees, instead of Engineering or Science, if they live in a neighbourhood that is governed by the ANC. It’s difficult to think why this would be: perhaps this confirms the old adage that it is not what you know but who you know. In economics-speak: social capital trumps human capital.

As we would expect, the quality of high school attended also matters. Says Bedasso: “High-achieving applicants who come from less competitive high schools tend to choose high-return majors than similar students from more competitive high schools.” So, if you’re from a poor school but do very well, you are more likely to study Math and Science than if you do equally well in a good school.

Surprisingly, whites weigh expected earnings more heavily in their choice of degree: “White applicants are more responsive to differentials in aptitude-adjusted expected earnings than black applicants.” In other words, whites are more likely to switch to a degree where they can earn a higher salary.

Bedasso thinks these results have profound implications for South Africa:

The gravitation of the children of the political elites towards less technical majors may deprive the political class of sufficient interest in productive activities. This, in turn, is likely to leave the elites with little incentives to respect property rights in the future. Hence, policy measures that will improve the availability of science education at high school level or account for the effect of near-peer role models in college admissions may go a long way in terms of shaping the path of economic development.

I think this is stretching the results, and would be more optimistic. Successful businesses require more than just breakthrough innovations; many of our top accountants and business students end up running technology companies because they know that running a business is not an algorithm to be programmed for success. As a new wave of young, successful black South African entrepreneurs strut their stuff in the business world and the barriers to entrepreneurial and managerial success appear less daunting, the attractiveness of a political career (and academic) will seem less appealing. This young generation, I would argue, is unlikely to cede property rights.

That being said, the need to promote math and science at schools remains imperative. We need more scientists, mathematicians, engineers, and computer programmers to remain competitive in the knowledge economy. We should start by appointing more and better math and science teachers, as Bedasso’s evidence suggests. It would also help if parents support their children to choose these (tougher) subjects. And if friends encourage each other.

I don’t know how to incentivize this change in behaviour. What I do know is that math and science can open doors that, if they remain locked, bar entry to a better life. Just ask a former prisoner from Benin.

Written by Johan Fourie

November 5, 2015 at 08:26