geographical economics

I love maps



I used to do economic geography research and still like to look at maps. People in the know will probably remember Adrian Firth’s dot maps of the 2011 census. They are still quite cool.

Currently, Facebook is putting together high resolution population maps, for among other countries South Africa, at 30 meter spatial resolution. The World Bank is working with this and has done some validation for Malawi. The Data Blog writes:

Facebook’s computer vision approach is a very fast method to produce spatially-explicit country-wide population estimates. Using their method, Facebook successfully generated at-scale, high-resolution insights on the distribution of buildings, unmatched by any other remote sensing effort to date. These maps demonstrate the value of artificial intelligence for filling data gaps and creating new datasets, and they could provide a promising complement to household surveys and censuses.

It is seriously cool.

Economics, links, Uncategorized

A few good links

The last week or two have been admin-intensive, but I stumbled across a few interesting bits along the way and want to share the links:

  • I attended a UNU-WIDER and National Treasury conference in Pretoria and enjoyed it a lot. The theme was “Growth and development policy – new data, new approaches and new evidence”. All that was from the different projects that have been using SARS administrative data for a firm-level look at employment, productivity, exporters etc. It was really interesting.
  • My old Warwick classmate Nick Powdthavee launched a new book on happiness research at the LSE and shared links on Facebook and Twitter. Here is a Vox.EU post on “The origins of happiness”: mental and physical health, and having a partner explains life satisfaction and then income matters…
  • And I caught a Chronicle post on “The personal lecture” with interesting ideas on big undergraduate groups, and on MOOCs.

Check it out.


Big data analytics

It is that dangerous time of the year when I am already done with 2016 (in my head) and start to think about things to do in 2017 with all the idealism of new year’s resolutions. Can an old dog economist still learn new big data analytics tricks?

So this is a note to self: Coursera has a course in R and Code School also has one that looks cool.

And then, a shout out to colleagues and their student featuring on the Development Impact blog: Martin Abel’s job market paper “The Value of Reference Letters”, is coauthored with Rulof Burger (SU) and Patrizio Piraino (UCT).


Random posts ahead!

After an academic year of teaching and also giving urgent feedback on templates it feels like I am finally surfacing after a deep dive, the sun is shining and I have ideas for blog posts again.

First, a recent video with a few thoughts of education and technology.

The point is that education technology can be very useful. However, we should not make voice-over-powerpoint video’s to replace the lecturer, but rather to amplify him/her. Being the lecturer is about much more than talking through the course content.

In this TED talk Ken Robinson makes the point much more eloquently.

#higher-ed, education, university

Professors as employees?

Yesterday Frances Woolley made an interesting post on the Worthwhile Canadian Initiative blog.

Prof employees

He quotes Akerlof and Kranton’s idea that a person who identifies with her employer – an insider – will think “she should work on behalf of the firm. Her ideal is to exert high effort.” He goes on to explain that there is often a misalignment between professorial effort and university priorities. Professors can at best be relied on to strive to be good professors, but we “are motivated by our disciplinary and professional identities, not our identities as university employees”.

This is something that managers frequently forget and they should read the whole post.

During these days of transformation, budget cuts and decolonialisation of the curriculum managers often talk about what “we” will do. We will improve the university’s standing in some or other rankings, we will pursue external consulting jobs to earn third stream income for the university, we will expand into Africa, or launch a MOOC.

It is going to take more than a few big ideas to get the Professors involved.


Policy and uncertainty

PUI koerant

This week I participated in the launch of the NWU Policy Uncertainty Index. The index number and report are the results of a small research project I have been working on with Prof Raymond Parsons. The press release is here on the School’s blog and we received a bit of media coverage as well in Beeld and Business Day.

One of the interesting questions we have been getting is how to distinguish between the impact of policy uncertainty and bad policy.



Good policy  ? ?
Bad policy  ?  ?

Our argument about uncertainty is the standard one: uncertainty is bad for planning, it raises the option value of waiting, it creates the possibility of fooling solutions etc. So, whether a policy is “good” or “bad”, uncertainty about it is always bad. It means that our index will always improve when there is greater certainty about good as well as bad policies – the key is that at least consumers and investors can plan and hedge.

Whether policies are good or bad depends very much on your point of view. I think that building Russian nuclear power stations is bad energy policy and the uncertainty about those plans makes for the worst possible outcome. The same can be said for the 50-50 ownership land reform proposals. It is more complicated when you think about last week’s monetary policy decision. Most Economists were certain that the SARB was going to increase rates but was that good or bad policy under the circumstances? The textbooks say that a central bank with an inflation target needs to show that they are committed to the target and build credibility, especially when it is painful to do so. Markets and ratings agencies are out there, reading the signals… But in this case the good monetary policy signal complicates matters for fiscal policy. The repo rate increase will dampen growth prospects further, lower the tax take and worsen the fiscal aggregates expressed as a percentage of GDP. And then the markets and rating agencies are worried all over again.

Does it help that we were certain about it? I would like to hear some thoughts.


HigherEd in the news

I cannot even try to join the discussion on higher education and #feesmustfall. It is moving too fast and there are some proper experts out there. I do want to share some links:

  • Johan and Co-Pierre  explained the opportunity cost of fees falling best in Business Day last year: Universities face an impossible trinity: appoint more black scholars, reduce student fees, or cut spending on outsourcing and maintenance of facilities. They argue that the bursary system should be reformed with targeted support of the students that need it.
  • In a Medium post Justin Goro argued for corporate support and a form of securitisation of future earnings.

I have wondered why no-one has mentioned the inequity of using more of the tax payer’s money to fund university students. Arguments for more public finding ignore the fact that higher education has some substantial private benefits (The Economist has U.S. data on the ‘value’ of university). Although there are poor students, they are not “the poor”.

We all know that education in South Africa faces many significant challenges (and if you want to learn more about it, follow Nic and read his blog), but it seems to me that giving spending more money on the 15 students that somehow had the resources to qualify to go to university, will only fuel further inequality. Where are the other 75 protesting?

Finally, you should read Shaun Stanley’s “Devil’s advocacy for decolonised  curricula” in the M&G Thought Leader. He argues that the way that the curricula is taught can disadvantage particular students. Food for thought.