Instead of publishing posts here I now publish all my writings on the open-access publication OurWorldInData.org.
I haven’t written much on this blog recently and so I thought it might make sense to just list – and link to – a couple of the projects that I’ve been working on during the last year. This way there are at least some things to read or watch for those who come to this page.
Our World in Data:
What kept me busy mostly was the work on Our World in Data, the free online publication on global development that I started some years ago.
Unfortunately quite a lot of time I spent on the search for funding. But there were also a lot of very positive developments!
- The publication grew quite a bit – we now have 87 entries on Our World in Data! You find them all listed on the landing page.
- Two of my favorite recent entries are:
- Jaiden Mispy, the web developer in our team, keeps making our own open source data visualization tool – the Our World in Data Grapher – more and more useful. Aibek Aldabergenov, our database developer, made it possible to access large development datasets directly (without uploading them manually) and that made the work of the authors much faster and more fun.
- And while I wasn’t active here on my personal blog, we actually now publish very regularly on the OWID-blog. My most widely read article in the last year is ‘The Short History of Living Conditions – And Why it Matters that We Know it’.
- But the biggest change for Our World in Data in the last year is for sure that we are now working as a team. And everyone in the team is really awesome and I’m really, really happy to be working with all of them!
My publications in the last year:
In addition to OWID I work as a researcher at the University of Oxford. These are the abstracts of the three papers I published in the last year:
Felix Pretis and Max Roser (2016) – Carbon Dioxide Emission-Intensity in Climate Projections: Comparing the Observational Record to Socio-Economic Scenarios. In Energy Volume 135, pages 718–725, September 2017. Available as an open-access publication at the journal’s website here.
Abstract: The large span of long-run projected temperature changes in climate projections does not predominately originate from uncertainty across climate models; instead it is the wide range of different global socio-economic scenarios and the implied energy production that results in high uncertainty about climate change. It is therefore important to assess the observational tracking of these scenarios. For the first time observations over two decades are available against which the initial sets of socio-economic scenarios used in IPCC reports can be assessed. Here we compare these socio-economic scenarios created in both 1992 and 2000 against the recent observational record to investigate the coupling of economic growth and fossil-fuel CO2 emissions. We find that the growth rate in fossil fuel CO2 emission intensity – fossil fuel CO2 emissions per GDP – over the 2000s exceeds the projections of all main emission scenarios. Proposing a method to disaggregate differences in global growth rates to country-by-country contributions, we find that the relative discrepancy is driven by high growth rates in Asia and Eastern Europe, in particular in Russia and China. The growth of emission intensity over the 2000s highlights the relevance of unforeseen local shifts in projections on a global scale.
Sterck, O., Roser, M., Ncube, M., Thewissen, S. (2017) – Allocation of development assistance for health: Is the predominance of national income justified?. In Health Policy and Planning – forthcoming.
Abstract: Major donors heavily rely on GNI per capita to allocate development assistance for health (DAH). This paper questions this paradigm by analyzing the determinants of health outcomes using cross-sectional data from 99 countries in 2012. We use disability-adjusted life years (Group I) per capita as our main indicator for health outcomes. We consider four primary variables: GNI per capita, institutional capacity, individual poverty and the epidemiological surroundings. We construct a health poverty line of 10·89 international-$ per day, which measures the minimum level of income an individual needs to have access to basic healthcare. We take the contagious nature of communicable diseases into account, by estimating the extent to which the population health in neighboring countries (the epidemiological surroundings) affects health outcomes. We apply a spatial two-stage least-squares model to mitigate the risks of reverse causality, and use additional IV estimations as sensitivity tests. Overall we find that GNI is not a significant predictor of health outcomes once other factors are controlled for.
Nolan, B., Roser, M., Thewissen, S. (2017) – GDP per capita versus median household income: What gives rise to divergence over time? In Review of Income and Wealth – forthcoming.
We published a summary of this paper here: Stagnating median incomes despite economic growth: Explaining the divergence in 27 OECD countries
Abstract: Divergence between the evolution of GDP per capita and the income of a “typical” household as measured in household surveys is giving rise to a range of serious concerns, especially in the USA. This paper investigates the extent of that divergence and the factors that contribute to it across 27 OECD countries, using data from OECD National Accounts and the Luxembourg Income Study. While GDP per capita has risen faster than median household income in most of these countries over the period these data cover, the size of that divergence varied very substantially, with the USA a clear outlier. The paper distinguishes a number of factors contributing to such a divergence, and finds wide variation across countries in the impact of the various factors. Further, both the extent of that divergence and the role of the various contributory factors vary widely over time for most of the countries studied. These findings have serious implications for the monitoring and assessment of changes in household incomes and living standards over time.
Together with Sir Tony Atkinson, Joe Hasell, and Salvatore Morelli, I published the new version of the Chartbook of Economic Inequality. The Chartbook aims to bring together the data on several dimensions of inequality for 25 countries over the period since 1900. Version 1 of this was published in 2014. And in May this year we published a major revision of this work.
And I published these two shorter articles in the last year:
- Why do we not hear the good news? – Washington Post; December 2016.
- The US lags far behind its peers on “inclusive” economic growth – with Stefan Thewissen; Vox; February 2017.
Collaborations in the last year:
With the great team of Kurz Gesagt I collaborated to make a short video that explains how rapid population happened – and why it will come to an end:
And in another video we told the history of the decline of maternal mortality:
Tony Atkinson & Hans Rosling
The very saddest changes in the last year were the deaths of Sir Tony Atkinson and Hans Rosling.
Tony died on the first day of this year. He brought me to Oxford, he helped to develop the ideas I am now working on, and he was just the kindest, smartest, and most generous mentor one could wish for. That was not only true for me, at his website I collected a long list of obituaries that speak for themselves.
Hans died only five weeks later than Tony. I definitely would not do what I am doing today if Hans had not done what he did. He had a massive impact on what I care about and what I spend my days working on. Personally he was exactly the same enthusiastic, witty, and clear-thinking man that he was on stage in his famous talks. For the British Medical Journal I wrote an obituary for him: ‘Seeing human lives in spreadsheets’ – Hans Rosling (1948–2017)
I miss both of them.
I’m part of the team that is building the web publication Our World In Data that presents the empirical evidence on global development.
We present this evidence topic by topic – war, democracy, child mortality, higher education, corruption and so on. One reason why I wanted all of these different aspects in one publication is that they are all related: better education for women leads to lower child mortality for example.
These cross-connections are really at the heart of development: from education to democracy, from democracy to war, from war to public debt, from public debt to spending to education etc. etc.
Because development has this structure we increasingly run into a problem however. The link between education for women and child mortality should be part of both entries – the one on education for women and the one on child mortality. And currently we have a similar section in both entries.
But an ideal system for such a publication might be to write that section X which discusses the link between A and B and then this same section X is shown in both entries, entry A and entry B. These modules would be reusable blocks – we write them once and use them several times. In computer science this is referred to as transclusion.
The idea for a ‘modular structure’, as I put in the title, would be to write these bits in modules:
- Education of women -> Child mortality
- Education -> Democracy
- Democracy -> Education
- Education -> Health of children
- Health of children -> Economic Growth
and then these modules are shown in the appropriate entries. Our current structure would work well since one module could always be headline-text-visualisation.
– This structure would us also help to possibly solve a second problem, namely that our entries get very long and they are becoming hard to navigate. A structure that breaks them in pieces in a good way might be helpful.
– It would then be straightforward to recombine modules to other ‘articles’. Every module that is tagged ‘USA’, or ‘long-term perspective’, or ‘public good finance’ etc.
My main questions is has anyone seen some structure for a web publication like this on WordPress? Are there helpful examples that we could look at?
ADDED ON 20 FEB: In response to the request above I’ve received several very helpful emails. Many thanks for this!
The source of this image and the background story is here (in Italian).
Average working hours in Belgium back then were 64 per week.
Trump is obsessed with comparisons of the US with China. In the last debate he said: “China is growing at 7 percent. And that for them is a catastrophically low number. We are growing — our last report came out — and it’s right around the 1 percent level. And I think it’s going down. … Look, our country is stagnant.”
This post is about why Trump’s comparison of the US with China is not useful. More broadly I want to defend democracies against the mistaken argument that authoritarian countries perform better economically than democracies.
Firstly, it is important to distinguish between two different forms of economic growth – growth at the technological frontier and catch up growth. Secondly, it is important to distinguish between the level of prosperity (GDP per capita) and the growth of prosperity (growth of GDP per capita).
If we don’t distinguish we get all confused about what political systems make possible or impossible for growth.
Growth at the technological frontier is very, very hard and always slow. The US was always at the frontier over the last 2 centuries and average growth never exceeded 2%. (in the last 2 years it was 1.5% and 1.8% – source.)
Catch-up growth is different from that, it is all about adopting existing technology and making use of it at a large scale. Catch-up is hard enough – if we knew how to do it everywhere we would have ended extreme poverty long ago. But once a country starts to catch up it can improve living conditions very rapidly; growth can be much higher than 2% for extended periods. And from what we have seen over the last decades autocratic rule is not necessarily an impediment to catch-up growth for very poor countries.
The confusion comes from the fact that richer countries are much much more often democratic countries. (The only countries who are rich and autocratic have an economy that is relying on the exports of fossil fuels.)
If we don’t distinguish between the two forms of growth we are in danger of comparing the fast growth in autocracies with the slow growth in democracies and take away that autocracies are better for growth.
The key to understanding why China and others could grow so fast is not that they are autocratic but that they are very, very poor. China’s level of GDP per capita is only a quarter of the American GDP per capita.
Fast catch-up growth can only happen in poor countries. What determines the “growth advantage” of autocracies is not that autocratic countries work better but that autocratic countries are poor.
Catch up growth can only happen in poor countries and it is very misleading to do what Trump does and to compare a country on the technological frontier (the US) with a very poor country (China) that is catching up. This is a same confusion that leads people to believe the ‘old argument’ above.
A fundamental problem in social science is that all good things come together. Developed countries are richer, healthier, happier, better educated, more democratic etc. The trouble for researchers – and ultimately policy makers – is then to find out what causes what. There is some research that suggests that democratic rule – ceteris paribus – is good for growth, but I would not think the evidence for this causal relationship is overwhelmingly clear.
What is however clear is that there is no reason to believe that a rich country at the technological frontier would do better if it was autocratically ruled. This false idea comes from the confusion of catch-up growth with growth at the frontier.
Don’t be fooled by China or Donald Trump, autocratic rule is not good for growth.