17 October 2014

Let them drown

3,000 people have drowned already this year trying to cross the Mediterranean to the EU, in pursuit of a better life. It is official UK government policy to not try and rescue such people, because that would only encourage others. I somehow find it hard to believe that even staunch opponents of immigration really think we should just stand by and watch people drown. 


via Duncan Stott and Phil Davis

27 September 2014

Good news from South Sudan

Charlie Goldsmith emails with updates on the Girl's Education South Sudan project:
"Our majority-South Sudanese team are proud that South Sudan, which has been so beset by trouble in the last year, has the chance to show positive ways in which it is a world-leader. 
Charlie Goldsmith Associates have been particularly involved on design, technology for, and delivery of: 
  • The South Sudan Schools Attendance Management System, through which enrolment and attendance of individual pupils – almost 900,000 of them by now – from top to bottom of the education system is recorded, with schools asked to report daily to a freephone number through SMSs from teachers’ own phones.
  • Cash Transfers to individual girls in P5-S4 and their families: more than 50,000 will be made in 2014, and around half a million, to 200,000 individual girls, by 2018. In 2015, we expect payment of the majority of these to be by M-Money. 
  • School capitation grants to fund investments in quality: almost 3000 schools have been approved to receive these grants, having passed hurdles including opening a bank account, and making a school development plan and budget, and there have been outstanding examples of value delivered, notably in terms of economical construction. GRSS is now looking at rolling this model of funding direct to service delivery units across to the health sector. 
  • A multi-year programme of investment in knowledge, evidence and research, much of it delivered by our specialist partners Forcier Consulting and, earlier on, Education for Change, including detailed school and household surveys, learning assessments, and a major subnational PFM performance survey.
The Government of the Republic of South Sudan (GRSS) budgets to spend 60m SSP of its own money a year (roughly £12m GBP) on school capitation grants, as part of the wider Local Services Support programme, supported by among others, the ODI Budget Strengthening Initiative. It's worth highlighting that this £12m a year spent by the government of South Sudan is slightly more than the DFID project itself (£60m over five and half years - a great example of leverage and sustainability where aid money can help to increase the effectiveness of potentially much larger government spending. 
You can find some examples of practical good news that has resulted – girls with the resources to be in school, teachers paid, classrooms and latrines quickly and cost-effectively constructed - and scans of how each school has budgeted to spend its capitation grant, including detail down to the cost of a latrine in a given school, and the accountability now being returned by schools.
In addition to the CGA work, other members of the consortium are delivering really exciting things too, particularly interesting are the BBC "Our School" programmes in local languages. Since they have one programme per State per topic, and have done about 14 topics this year, it is really quite a significant library of decent resources of people saying sensible things about education, including the practicalities, not just the slogans, in their own language. In due course, the project is going to do important things on in classroom education quality too. 
Using innovative technology elsewhere

We think some of these approaches and tools are broadly applicable:
  • Monitoring enrolment and attendance at an individual level, in near-real time, on a public website (data on individuals can only be accessed by permissioned login), is a step of assurance and usefulness of data that goes beyond what some EMISs offer.
  • Getting funds direct into the hands of individuals and down to bank accounts of service delivery units like schools and primary health care units, gives country governments and their partners assurance that funds have reached their destination, and sets good incentives for funds to be used correctly, and the leverage and information to follow up if there is a problem. Putting funds to individuals and schools and clinics at local level stimulates the local economy, and, in particular, financial inclusion – in South Sudan, Eden Commercial Bank are opening five new branches in County towns during 2014, driven partly by the additional transaction volume provided by capitation grants and cash transfers, and other banks have set up travelling account opening services. 
  • The technology approach is designed for a low-connectivity/not-always-on environment, and for users using their own mobile phones/devices.
These approaches are operationally effective, and showing promising signs of effect on retention and enrolment, in South Sudan: how much more might they achieve in an environment where there were fewer barriers to accessing public services?"

15 July 2014

This is why I don't care about climate change

Well, not "don't care at all", but, you know, not as much as about poverty and development. Stefan Dercon puts it better than I ever have:
Poverty reduction tends to be strongly linked to economic growth, but growth impacts the environment and increases CO2 emissions. So can greener growth that is more climate-resilient and less environmentally damaging deliver large scale poverty reduction? ... We argue that there are bound to be trade-offs between emissions reductions and a greener growth on the one hand, and growth that is most effective in poverty reduction. We argue that development aid, earmarked for the poorest countries, should only selectively pay attention to climate change, and remain focused on fighting current poverty reduction, including via economic growth, not least as future resilience of these countries and their population will depend on their ability to create wealth and build up human capital now. The only use for development aid within the poorest countries for explicit climate-related investment ought to be when the investments also contribute to poverty reduction now

Value for money in technical assistance to governments

The DFID project completion report is out (here) for the South Sudan ODI fellows from 2009-2012. It's pretty good. (this doesn't include my cohort).
the fellows delivered – and exceeded - the desired outputs and the programme has achieved – and exceeded – the desired outcome, at slightly under budget. Given the minimal oversight given to this programme by DFID South Sudan, a large part of the credit must go to the project partner, ODI, at least in respect of its selection and briefing of the fellows, who were very well suited to the tasks in hand. The majority of credit must, however, go to the fellows themselves, for undertaking their work professionally and working to sustainably build colleagues’ skills and capacity. 
Taking into account all of the evidence gathered in this review it seems clear that the ODI Fellowship for the Government of Southern Sudan programme delivered very strong VFM over the review period 2009 – 2012. It is an impressively performing programme, particularly given the difficult context to deliver results in South Sudan that it managed to overcome - if anything performing better in value for money terms than the global ODI programme did in more benign environments. 
This review found that the programme was implemented to expected timelines and budgets, with strong performance by the fellows translating into very strong performance on value for money metrics. The programme over-achieved in relation to the desired outputs and outcome, while making a small cost saving.

27 June 2014

How not to improve education in India

Some great analysis from MINT who highlight a new Government of India report, which ranks state education "outcomes".

What is odd is that the government rank has a negative correlation with the rankings of the Pratham report which directly measures learning outcomes.


So what goes into the government "outcomes" index?

- Number of teaching days
- Teacher working hours
- Enrolment rates
- Drop-out rates
- Primary-to-secondary transition rates

These are all basically inputs, with the exception of drop-outs and transition rates, which maybe say something about quality. But none of them are actually directly measuring learning at all. Yet more evidence for the Lant Pritchett case that focusing on inputs or "EMIS-visible" metrics won't get us quality learning outcomes, and measuring learning directly is critical to focusing policy attention on how to improve learning.

HT: Abhijeet Singh

12 June 2014

Don't shit on your own doorstep

I was talking to a water and sanitation programme manager a few weeks ago, who seemed frustrated that these stupid people kept crapping everywhere. Why would you shit on your own doorstep? The programme had several "behaviour change" interventions (horrible phrase, slightly Orwellian no?), but really, how hard should it be to not shit in the open?

One of the great things about economics is that it does not assume that people are just being dumb. It treats people with respect, and assumes first that there is probably a good reason why they are doing something which might seem irrational. I don't really know enough about water and sanitation, but I was suspicious of the idea that these recalcitrant natives just couldn't figure out what was good for them.

Does this paper prove me right?
"latrine use constitutes an externality rather than a pure private gain: It is the open defecation of one’s neighbors, rather than the household’s own practice, that matters most for child survival. The gradient and mechanism we uncover have important implications for child health and mortality worldwide, since 15% of the world’s population defecates in the open. To put the results in context, we find that moving from a locality where everybody defecates in the open to a locality where nobody defecates in the open is associated with a larger difference in child mortality than moving from the bottom quintile of asset wealth to the top quintile of asset wealth."
The problem then is a "simple" collective action problem (simple in the sense of understanding the nature of the problem, not at all simple to solve). This isn't that complicated stuff.

HT: kim yi dionne

03 June 2014

Important Research Funding Opportunities: Quantifying the economic impact of Shakira (for UNICEF)

"Shakira Mebarak, world-famous singer and songwriter, is a devoted advocate for children. The singer, known professionally as Shakira, was appointed a UNICEF Goodwill Ambassador on 24 October 2003."

Which is all very well and everything, but surely what we all really care about is exactly how valuable is she to the UNICEF marketing team? And what is it about her that makes her valuable? Are singers more or less valuable than actresses? Blondes or brunettes? Men or women? Does fundraising value depreciate over time with age? If you've always been fascinated by important research questions like these, well do UNICEF have the RFP for you (sadly I think the quantitative celebrity fundraising research and analytics team at my office is busy right now, so I'm generously passing this on. You're welcome). Thank god UNICEF is taking evidence-based decision-making seriously where it really matters.

"Quantitative Research: Identifying the Right Celebrities for UNICEF Partnerships & Public Attitudes towards Celebrity Partnerships 
The purpose of this Request for Proposal (RFP) is to seek proposals from qualified agencies to provide quantitative research (using System 1 approach) for identifying the right celebrities for UNICEF partnerships & Public Attitudes towards Celebrity Partnerships."



Why governments don't like private schools?

Here are a few excerpts from the new textbook delivered to millions of primary school children in Venezuela:
1. The first page of each [book] starts with the words “Hugo Chavez: Supreme Commander of the Bolivarian Revolution.”  
2. They describe Chavez as the man who liberated Venezuela from tyranny, at times making him appear more important than 19th century founding father Simon Bolivar.  
3. The books present a 2002 coup that briefly toppled Chavez as an insurrection planned by Washington while playing down the role of massive opposition protests in this deeply divided country.

28 May 2014

Does growth always depend on natural resources?

George Monbiot writes:
"Economic growth is an artefact of the use of fossil fuels."
Is this right? Actually most economists think that growth is driven by ideas and innovation not raw inputs. This 3 minute video by Deidre McCloskey provides a short economic history of growth since the beginning of time.



Can this really be true? Below are two charts showing energy consumption and GDP per capita from 1970 to 2012. In the UK, whilst our per capita GDP has doubled, our energy consumption has barely moved, and actually slightly decreased.

Of course there is a caveat to this story, and its a pretty big caveat. Whilst growth at the technological frontier (in advanced economies) can only be driven by innovation, in developing countries further away from the frontier, catch-up growth is possible by pure investment and copying existing technologies (such as, er coal-fired power plants). So whilst rich countries don't necessarily need to increase their overall energy consumption to grow, developing countries almost certainly do.

So the question for rich environmentalists is: was there one rule for us and another for everyone else? Or is it actually incumbent on us to invent some better technologies for the world to copy, rather than expecting them to choose between polluting the environment (like we did) or continuing to live on $3 a day for the rest of their lives?

08 May 2014

DevBalls - Exposing the absurdities of the aid industry

Someone at DFID seems to have had enough.
"DevBalls is an online space for comment on the international development aid industry. 
DevBalls is here because the aid industry has – functionally and morally – lost its way. And those who should hold it to account - the media, researchers, politicians - don’t. DevBalls is here because aid can only become better when its absurdities and hypocrisies are open to view. 
DevBalls is compiled by a group of aid professionals who control its content. We welcome relevant contributions sent to DevBalls1@gmail.com. Anonymity is guaranteed."

The blogosphere has been pretty light on cyncial scorn since the demise of Bill Easterly's AidWatch, so DevBalls looks like one to watch. Won't be comfortable reading for DFID or ODI or ASI.

16 April 2014

The origins of "the dismal science"

"This viewpoint infuriates some critics of economics, to the extent that it earned the famous nickname of “the dismal science”. Too few people know the context in which Thomas Carlyle hurled that epithet: it was in a proslavery article, first published in 1849, a few years after slavery had been abolished in the British empire. Carlyle attacked the idea that “black men” might simply be induced to work for pay, according to what he sneeringly termed the “science of supply and demand”. Scorning the liberal views of economists, he believed Africans should be put to work by force."
That's Tim Harford.

Wikipedia has more:
"However, the full phrase "the dismal science" first occurs in Carlyle's 1849 tract entitled Occasional Discourse on the Negro Question, in which he was arguing for the reintroduction of slavery as a means to regulate the labor market in the West Indies
Not a "gay science," I should say, like some we have heard of; no, a dreary, desolate and, indeed, quite abject and distressing one; what we might call, by way of eminence, the dismal science. 
It was "dismal" in "find[ing] the secret of this Universe in 'supply and demand,' and reducing the duty of human governors to that of letting men alone." Instead, the "idle Black man in the West Indies" should be "compelled to work as he was fit, and to do the Maker's will who had constructed him."[4]"
In which case I'm proud to be dismal.  

15 April 2014

Do teachers skip class because of low pay?

Teacher absenteeism is a huge problem in developing countries, wasting up to a quarter of all spending on primary education in developing countries.

The 2014 Education for All Global Monitoring Report, which was launched in London last week, puts the problem mainly down to the low pay and poor working conditions of teachers.
"While teacher absenteeism and engagement in private tuition are real problems, policy-makers often ignore underlying reasons such as low pay and a lack of career opportunities. ...  Policy-makers need to understand why teachers miss school. In some countries, teachers are absent because their pay is extremely low, in others because working conditions are poor. In Malawi, where teachers’ pay is low and payment often erratic, 1 in 10 teachers stated that they were frequently absent from school in connection with financial concerns, such as travelling to collect salaries or dealing with loan payments. High rates of HIV/AIDS can take their toll on teacher attendance."
The report includes this chart, showing that in a handful of countries teachers earn below $10 a day (which they have decided is not enough to live on).


Which seems jarring when the same week there was a conference on the economics of education in developing countries, where much of the literature is focused exactly on this issue of teacher absenteeism, and finds very little evidence that low pay is the main factor (as opposed to, say, weak or non-existent systems of accountability). In India it is well documented that whereas absenteeism is roughly similar in public and private schools, teachers in public schools are paid more than 5 times as much as private school teachers.

(See for example this chart from data from Singh 2013, or similar from Kremer et al 2005Alcazar et al 2006 in Peru, or African data here)


Harry Patrinos of the World Bank writes:
"There is very little evidence that higher salaries lead to better attendance, however. Contract teachers have the same or higher absence rates. Compared to public school teachers, though, private school teachers are absent less, even though contract and private school teachers alike take home much less pay than their regular civil service public school teacher counterparts."
As little as teachers might make in some countries, they are still doing well relative to most other people. In many countries public primary school teachers are the 1%.

I thought I'd take a quick look at the data presented in the GMR and see what those teacher salaries are presented as a % of GDP. In OECD countries, average teacher salary is roughly around the same level as GDP per capita. In African countries, the average teacher salary is 3 - 4 times GDP per capita.


Karthik Muralidharan summarised the state of public schools in India as facing two problems; governance and pedagogy. This probably generalises to much of the developing world. What this GMR comes across as doing is focusing almost entirely on the pedagogy problem, and sweeping the governance problem under the carpet (receiving roughly 10 pages attention out of a 300 page report). Perhaps this is a welcome counterbalance to prominent World Bank research which focuses much more on the governance problem. But really shouldn't a major flagship state of the sector report aspire to properly tackle both? Of course fixing the pedagogy problem means working with teachers to improve their capabilities and not demonising them or calling them all lazy slackers. But neither can we just ignore the reality of skiving on a massive scale (or: Don't hate the player, hate the game). 

14 April 2014

You won't believe these 8,000 children who are actually going to starve to death today

I'm trying to write a pithy summary or pick a smart quote from Abhijeet Singh's new blog about malnutrition up on Ideas for India but it's hard not to just be deeply depressed when thinking about malnutrition. We apparently live in the 21st Century where flying robots and self-driving cars are real things, yet we aren't collectively bothered enough to do anything about the 8,000 children who starve to death every single day (three million a year). And that's partly because as humans we're more interested in what is interesting than what is true or what is important. 8,000 children starving to death everyday is just something that happens. It isn't new or counterintuitive or surprising.

So Abhijeet's paper is interesting and tells us something different, which should be applauded really just for finding a new angle to bring some attention to one of the most important but dull outrageous injustices there are. The conventional wisdom is that stunting in the first thousand days of life is irreversible. Abhijeet presents evidence to the contrary that giving children a meal every day at age 5 can fully make up for malnutrition due to a drought at age 1. So the policy conclusion is what - don't write-off malnourished children after a thousand days? Or how about maybe how on earth are we still letting children starve in the first place? Enjoy your lunch.